Merge branch 'main' into pdf-text-editor-v2-pdfium-2

2026-09-02 21:03:34 +03:00 · 2026-08-24 22:20:57 +01:00
parent 5eb267bd7b e51885b379
commit 448f45ad96
152 changed files with 2469 additions and 2512 deletions
@@ -96,6 +96,14 @@ configs/
 __pycache__/
 **/__pycache__/

+# Python virtualenvs. Large, platform-specific, and their symlinks break the build.
+.venv/
+**/.venv/
+venv/
+**/venv/
+*.egg-info/
+**/*.egg-info/
+
 # Local env
 .env
 .env.*
@@ -22,26 +22,15 @@ indent_size = 4

 [*.html]
 indent_size = 2
-insert_final_newline = false
-trim_trailing_whitespace = false

-[{*.js,*.jsx,*.mjs,*.ts,*.tsx}]
+[{*.js,*.jsx,*.mjs,*.ts,*.tsx,*.mts}]
 indent_size = 2

 [*.css]
-# CSS files typically use an indent size of 2 spaces for better readability and alignment with community standards.
 indent_size = 2

 [*.{yml,yaml}]
-# YAML files use an indent size of 2 spaces to maintain consistency with common YAML formatting practices.
-indent_size = 2
-insert_final_newline = false
-trim_trailing_whitespace = false
-
-[*.json]
-# JSON files use an indent size of 2 spaces, which is the standard for JSON formatting.
 indent_size = 2

-[*.jsonc]
-# JSONC (JSON with comments) files also follow the standard JSON formatting with an indent size of 2 spaces.
+[*.{json,jsonc}]
 indent_size = 2
@@ -39,7 +39,7 @@ jobs:
      pr_ref: ${{ steps.resolve.outputs.ref }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -121,7 +121,7 @@ jobs:

    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -475,7 +475,7 @@ jobs:

    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -70,7 +70,7 @@ jobs:
      enable_prototypes: ${{ steps.check-prototypes-flag.outputs.enable_prototypes }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -177,7 +177,7 @@ jobs:

    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -482,7 +482,7 @@ jobs:
      issues: write # add/remove labels, delete the command comment
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -22,7 +22,7 @@ jobs:

    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -38,7 +38,7 @@ jobs:
      is_fork: ${{ steps.decide.outputs.is_fork }}
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -20,7 +20,7 @@ jobs:
      pull-requests: write
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -26,7 +26,7 @@ jobs:
      jar_sha256: ${{ steps.hashes.outputs.jar_sha256 }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -71,7 +71,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -18,7 +18,7 @@ jobs:
      issues: write # labels are applied through the issues API
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -31,7 +31,7 @@ jobs:
        flavor: [core, proprietary, saas]
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -58,7 +58,7 @@ jobs:
      SYSTEM_ENABLEANALYTICS: "false"
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -351,7 +351,7 @@ jobs:
      MN_COMPOSE: docker-compose-multinode.yml
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -49,7 +49,7 @@ jobs:
      proprietary: ${{ steps.changes.outputs.proprietary }}
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -298,7 +298,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -28,7 +28,7 @@ jobs:
      pull-requests: write
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -16,7 +16,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -17,7 +17,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -28,7 +28,7 @@ jobs:
      pull-requests: write # Allow writing to pull requests
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -34,7 +34,7 @@ jobs:
    timeout-minutes: 15
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -20,7 +20,7 @@ jobs:
    timeout-minutes: 30
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -14,7 +14,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -28,7 +28,7 @@ jobs:

    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -18,7 +18,7 @@ jobs:
    timeout-minutes: 30
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -32,7 +32,7 @@ jobs:
            project: stubbed-webkit
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -21,7 +21,7 @@ jobs:
    timeout-minutes: 25
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -28,7 +28,7 @@ jobs:
      licenses-backend: ${{ steps.changes.outputs.licenses-backend }}
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -56,7 +56,7 @@ jobs:
      repository-projects: write # Required for enabling automerge
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -334,7 +334,7 @@ jobs:
      repository-projects: write # Required for enabling automerge
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -15,7 +15,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -18,7 +18,7 @@ jobs:
    timeout-minutes: 15
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
      - name: Checkout repository
@@ -15,7 +15,7 @@ jobs:
      issues: write
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -48,7 +48,7 @@ jobs:
      version: ${{ steps.versionNumber.outputs.versionNumber }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -140,7 +140,7 @@ jobs:
            file_suffix: "-server"
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -207,7 +207,7 @@ jobs:
      RELEASE_GPG_PRIVATE_KEY: ${{ secrets.RELEASE_GPG_PRIVATE_KEY }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit
          allowed-endpoints: >
@@ -715,7 +715,7 @@ jobs:
      contents: write
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -23,7 +23,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -74,7 +74,7 @@ jobs:
    timeout-minutes: 60
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -140,7 +140,7 @@ jobs:
      contents: read
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -28,7 +28,7 @@ jobs:
      jar_sha256: ${{ steps.hashes.outputs.jar_sha256 }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -80,7 +80,7 @@ jobs:
      contents: write
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -32,7 +32,7 @@ jobs:
      pull-requests: write # pulls.get/list plus add/remove the label on PRs
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -14,7 +14,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -48,7 +48,7 @@ jobs:
          echo "version=${VERSION}" >> $GITHUB_OUTPUT

      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -18,6 +18,16 @@ on:
        required: false
        type: boolean
        default: false
+      build_engine:
+        description: "Build & push the standalone stirling-engine image."
+        required: false
+        type: boolean
+        default: true
+      force_engine_rebuild:
+        description: "Rebuild stirling-engine even if its source hash is unchanged."
+        required: false
+        type: boolean
+        default: false
  push:
    branches:
      - release
@@ -50,9 +60,10 @@ jobs:
    env:
      RUN_MAIN_APP: ${{ github.event_name != 'workflow_dispatch' || inputs.build_main_app }}
      RUN_UNOSERVER: ${{ github.event_name != 'workflow_dispatch' || inputs.build_unoserver }}
+      RUN_ENGINE: ${{ github.event_name != 'workflow_dispatch' || inputs.build_engine }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -387,3 +398,119 @@ jobs:
          else
            echo "Warning: COSIGN_PRIVATE_KEY not set, skipping unoserver image signing"
          fi
+
+      # Standalone AI engine image, same shape as the unoserver image above.
+      - name: Compute engine image source hash
+        id: engineHash
+        if: env.RUN_ENGINE == 'true'
+        run: |
+          set -eu
+          hash=$( { cat engine/Dockerfile engine/pyproject.toml engine/uv.lock engine/.env; \
+                    find engine/src -type f -print0 | sort -z | xargs -0 cat; } \
+                  | sha256sum | cut -d' ' -f1)
+          echo "hash=${hash}" >> "$GITHUB_OUTPUT"
+          echo "Engine source hash: ${hash}"
+
+      - name: Decide whether to publish engine image
+        id: engineDecision
+        if: env.RUN_ENGINE == 'true'
+        env:
+          ENGINE_VERSION: ${{ steps.versionNumber.outputs.versionNumber }}
+          ENGINE_HASH: ${{ steps.engineHash.outputs.hash }}
+          ENGINE_IMAGE: ghcr.io/${{ steps.repoowner.outputs.lowercase }}/stirling-engine
+          ENGINE_HASH_ANNOTATION: org.stirlingpdf.engine-source-hash
+          FORCE_REBUILD: ${{ inputs.force_engine_rebuild }}
+          GH_REF: ${{ github.ref }}
+          EVENT_NAME: ${{ github.event_name }}
+        run: |
+          set -eu
+          mode="skip"
+          tags=""
+
+          read_published_hash() {
+            local ref="$1"
+            docker buildx imagetools inspect "$ref" --raw 2>/dev/null \
+              | jq -r --arg key "$ENGINE_HASH_ANNOTATION" \
+                  '.annotations[$key] // empty' \
+              2>/dev/null || true
+          }
+
+          # Manual dispatch from any branch routes to the :alpha publish path.
+          EFFECTIVE_REF="$GH_REF"
+          if [ "$EVENT_NAME" = "workflow_dispatch" ]; then
+            EFFECTIVE_REF="refs/heads/testMain"
+          fi
+
+          case "$EFFECTIVE_REF" in
+            refs/heads/release)
+              if [ "${FORCE_REBUILD}" = "true" ]; then
+                echo "force_engine_rebuild=true — building stable regardless"
+                mode="stable"
+                tags="${ENGINE_IMAGE}:${ENGINE_VERSION},${ENGINE_IMAGE}:latest"
+              elif docker manifest inspect "${ENGINE_IMAGE}:${ENGINE_VERSION}" >/dev/null 2>&1; then
+                echo "stirling-engine:${ENGINE_VERSION} already on GHCR — skipping"
+              else
+                echo "stirling-engine:${ENGINE_VERSION} is new — will publish"
+                mode="stable"
+                tags="${ENGINE_IMAGE}:${ENGINE_VERSION},${ENGINE_IMAGE}:latest"
+              fi
+              ;;
+            refs/heads/main|refs/heads/testMain)
+              published_hash=$(read_published_hash "${ENGINE_IMAGE}:alpha")
+              if [ "${FORCE_REBUILD}" = "true" ]; then
+                echo "force_engine_rebuild=true — rebuilding :alpha regardless"
+                mode="alpha"
+                tags="${ENGINE_IMAGE}:alpha"
+              elif [ -n "$published_hash" ] && [ "$published_hash" = "$ENGINE_HASH" ]; then
+                echo "Published :alpha source hash matches (${published_hash}) — skipping"
+              else
+                if [ -z "$published_hash" ]; then
+                  echo ":alpha has no source-hash annotation (first publish) — will publish"
+                else
+                  echo "Source hash changed (was ${published_hash}, now ${ENGINE_HASH}) — will publish"
+                fi
+                mode="alpha"
+                tags="${ENGINE_IMAGE}:alpha"
+              fi
+              ;;
+            *)
+              echo "Branch ${GH_REF} does not publish engine image"
+              ;;
+          esac
+          echo "mode=${mode}" >> "$GITHUB_OUTPUT"
+          echo "tags=${tags}" >> "$GITHUB_OUTPUT"
+
+      - name: Build and push engine image
+        id: build-push-engine
+        if: env.RUN_ENGINE == 'true' && steps.engineDecision.outputs.mode != 'skip'
+        uses: docker/build-push-action@53b7df96c91f9c12dcc8a07bcb9ccacbed38856a # v7.3.0
+        with:
+          builder: ${{ steps.buildx.outputs.name }}
+          context: .
+          file: ./engine/Dockerfile
+          push: true
+          cache-from: type=gha,scope=stirling-engine
+          cache-to: type=gha,mode=max,scope=stirling-engine
+          tags: ${{ steps.engineDecision.outputs.tags }}
+          # Manifest annotation read by the decision step above to detect drift.
+          annotations: |
+            index:org.stirlingpdf.engine-source-hash=${{ steps.engineHash.outputs.hash }}
+          platforms: linux/amd64,linux/arm64/v8
+          provenance: true
+          sbom: true
+
+      - name: Sign engine image
+        if: env.RUN_ENGINE == 'true' && steps.engineDecision.outputs.mode == 'stable'
+        env:
+          DIGEST: ${{ steps.build-push-engine.outputs.digest }}
+          TAGS: ${{ steps.engineDecision.outputs.tags }}
+          COSIGN_PRIVATE_KEY: ${{ secrets.COSIGN_PRIVATE_KEY }}
+          COSIGN_PASSWORD: ${{ secrets.COSIGN_PASSWORD }}
+        run: |
+          if [ -n "$COSIGN_PRIVATE_KEY" ]; then
+            echo "$TAGS" | tr ',' '\n' | while read -r tag; do
+              cosign sign --key env://COSIGN_PRIVATE_KEY --yes "${tag}@${DIGEST}"
+            done
+          else
+            echo "Warning: COSIGN_PRIVATE_KEY not set, skipping engine image signing"
+          fi
@@ -19,7 +19,7 @@ jobs:
      packages: write
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -35,7 +35,7 @@ jobs:

    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -17,7 +17,7 @@ jobs:
      pull-requests: write
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -30,7 +30,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -33,7 +33,7 @@ jobs:
      pull-requests: write
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -37,7 +37,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -72,7 +72,7 @@ jobs:
      matrix: ${{ steps.set-matrix.outputs.matrix }}
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -134,7 +134,7 @@ jobs:
      SIGN_BUNDLE: ${{ inputs.sign && (matrix.platform == 'macos-15' && secrets.APPLE_CERTIFICATE != '' || github.ref == 'refs/heads/main') }}
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -703,7 +703,7 @@ jobs:
      pull-requests: write
    steps:
      - name: Harden the runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -795,7 +795,7 @@ jobs:
    if: always()
    steps:
      - name: Harden the runner (Audit all outbound calls)
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -34,7 +34,7 @@ jobs:
    timeout-minutes: 20
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -77,7 +77,7 @@ jobs:
            cache-scope: stirling-pdf-fat
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -217,7 +217,7 @@ jobs:
    runs-on: ubuntu-latest
    steps:
      - name: Harden Runner
-        uses: step-security/harden-runner@b09bb98e06d4d774595224525879c09bc6e98c40 # v2.20.1
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -19,7 +19,7 @@ jobs:
    timeout-minutes: 20
    steps:
      - name: Harden runner
-        uses: step-security/harden-runner@bf7454d06d71f1098171f2acdf0cd4708d7b5920 # v2.20.0
+        uses: step-security/harden-runner@05e31511f85b41b11d1cf0ef85d0992719546e2c # v2.21.0
        with:
          egress-policy: audit

@@ -375,13 +375,13 @@ tasks:
    desc: "Auto-fix code formatting"
    deps: [install]
    cmds:
-      - npx prettier --write .
+      - npx oxfmt --write .

  format:check:
    desc: "Check code formatting"
    deps: [install]
    cmds:
-      - npx prettier --check .
+      - npx oxfmt --check .

  fix:
    desc: "Auto-fix lint and format"
@@ -554,6 +554,7 @@ tasks:
    deps: [install, ":backend:swagger"]
    cmds:
      - npx tsx editor/scripts/generate-tool-api-types.mts --spec ../SwaggerDoc.json --output editor/src/core/types/toolApiTypes.ts --io-output editor/src/core/types/toolIO.ts
+      - task: format
    sources:
      - editor/scripts/generate-tool-api-types.mts
      - ../SwaggerDoc.json
@@ -563,9 +564,9 @@ tasks:

  tool-models:check:
    desc: "Fail if committed tool API types are out of date"
-    deps: [install, ":backend:swagger"]
    cmds:
-      - npx tsx editor/scripts/generate-tool-api-types.mts --spec ../SwaggerDoc.json --output editor/src/core/types/toolApiTypes.ts --io-output editor/src/core/types/toolIO.ts --check
+      - task: tool-models
+      - git diff --exit-code -- editor/src/core/types/toolApiTypes.ts editor/src/core/types/toolIO.ts

  licenses:generate:
    desc: "Generate frontend license report"
@@ -18,7 +18,7 @@ dependencies {
    api "org.apache.pdfbox:preflight:$pdfboxVersion"
    api 'com.github.junrar:junrar:8.0.0' // RAR archive support for CBR files
    api 'jakarta.servlet:jakarta.servlet-api:6.1.0'
-    api 'org.snakeyaml:snakeyaml-engine:3.0.1'
+    api 'org.snakeyaml:snakeyaml-engine:3.1.1'
    api "org.springdoc:springdoc-openapi-starter-webmvc-ui:3.0.3"
    // Simple Java Mail for EML/MSG parsing (replaces direct Angus Mail usage)
    api 'org.simplejavamail:simple-java-mail:9.3.2'
@@ -91,24 +91,32 @@ public class ControllerAuditAspect {
        MethodSignature sig = (MethodSignature) joinPoint.getSignature();
        Method method = sig.getMethod();

-        // Fast path: check if auditing is enabled before doing any work
-        // This avoids all data collection if auditing is disabled
-        if (!auditService.shouldAudit(method, auditConfig)) {
+        // Resolve the event type up front so the enterprise gate can be type-aware: document
+        // processing events (the Documents tab's data source) are audited without an Enterprise
+        // license, while the rest of the audit log stays Enterprise-only. resolveEventType is cheap
+        // (annotation / class / path checks), so it's safe on the pre-record fast path.
+        Audited auditedAnnotation = method.getAnnotation(Audited.class);
+        String path = getRequestPath(method, httpMethod);
+        AuditEventType eventType =
+                auditService.resolveEventType(
+                        method,
+                        joinPoint.getTarget().getClass(),
+                        path,
+                        httpMethod,
+                        auditedAnnotation);
+
+        // Fast path: skip all data collection when this event won't be recorded.
+        if (!auditService.shouldAudit(eventType, method, auditConfig)) {
            return joinPoint.proceed();
        }

-        // Check if method is explicitly annotated with @Audited
-        Audited auditedAnnotation = method.getAnnotation(Audited.class);
        AuditLevel level = auditConfig.getAuditLevel();
-
        // If @Audited annotation is present, respect its level setting
        if (auditedAnnotation != null) {
            // Use the level from annotation if it's stricter than global level
            level = auditedAnnotation.level();
        }

-        String path = getRequestPath(method, httpMethod);
-
        // Skip static GET resources
        if ("GET".equals(httpMethod)) {
            HttpServletRequest maybe = auditService.getCurrentRequest();
@@ -209,15 +217,6 @@ public class ControllerAuditAspect {
                // the body ran, so it must happen here rather than with the pre-proceed HTTP data).
                auditService.addAutomationContext(data, req);

-                // Resolve the event type using the unified method
-                AuditEventType eventType =
-                        auditService.resolveEventType(
-                                method,
-                                joinPoint.getTarget().getClass(),
-                                path,
-                                httpMethod,
-                                auditedAnnotation);
-
                // Add result only if operation result capture is explicitly enabled
                // Skip result for UI_DATA events to avoid storing large response bodies
                if (auditService.shouldCaptureOperationResults()
@@ -0,0 +1,13 @@
+package stirling.software.proprietary.audit;
+
+import org.springframework.stereotype.Component;
+
+/** Self-hosted default: any portal user sees the whole-server documents queue. */
+@Component
+public class DefaultPortalDocumentsScopeResolver implements PortalDocumentsScopeResolver {
+
+    @Override
+    public PortalAuditScope resolve() {
+        return PortalAuditScope.server();
+    }
+}
@@ -0,0 +1,7 @@
+package stirling.software.proprietary.audit;
+
+/** Resolves which slice of the documents queue a portal user may see. */
+public interface PortalDocumentsScopeResolver {
+
+    PortalAuditScope resolve();
+}
@@ -9,6 +9,7 @@ import java.util.Map;
 import java.util.Set;

 import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDDocumentInformation;
 import org.springframework.beans.factory.annotation.Autowired;
 import org.springframework.core.io.Resource;
 import org.springframework.http.MediaType;
@@ -20,12 +21,13 @@ import org.springframework.web.bind.annotation.RestController;
 import org.springframework.web.multipart.MultipartFile;

 import io.github.pixee.security.Filenames;
-import io.swagger.v3.oas.annotations.Hidden;
 import io.swagger.v3.oas.annotations.Operation;
 import io.swagger.v3.oas.annotations.tags.Tag;

 import lombok.extern.slf4j.Slf4j;

+import stirling.software.common.model.tool.ToolFormat;
+import stirling.software.common.model.tool.ToolIO;
 import stirling.software.common.service.CustomPDFDocumentFactory;
 import stirling.software.common.service.PdfMetadataService;
 import stirling.software.common.service.UserServiceInterface;
@@ -48,11 +50,13 @@ import tools.jackson.databind.node.ObjectNode;
 * <p>Runs as a Classification-policy pipeline step: it reads a bounded page window, asks the AI
 * engine to classify the document against the built-in label set, and stores the engine's JSON
 * answer — minus the transport-only {@code outcome} field — in the custom Info-dictionary key
- * {@link PdfMetadataService#CLASSIFICATION_KEY}. Returns the labelled PDF. Not intended for direct
- * client use.
+ * {@link PdfMetadataService#CLASSIFICATION_KEY}. Returns the labelled PDF.
+ *
+ * <p>Published in the API spec rather than hidden, so the tool-model generator emits it and a
+ * pipeline can name it as a step like any other tool. Classification is a thing a pipeline does,
+ * not a thing only the Classification policy may do.
 */
@Slf4j
-@Hidden
@RestController
@RequestMapping("/api/v1/ai/tools")
@Tag(name = "AI Tools", description = "Dispatchable AI-backed tools.")
@@ -99,19 +103,31 @@ public class ClassifyLabelController {
    }

    @PostMapping(value = "/classify-and-label", consumes = MediaType.MULTIPART_FORM_DATA_VALUE)
+    // PDF in, the same PDF out with a verdict on it, so a chain can be checked across this step.
+    @ToolIO(accepts = ToolFormat.PDF, produces = ToolFormat.PDF)
    @Operation(
            summary = "Classify a PDF and label its metadata",
            description =
                    "Reads the first two and last two pages, classifies the document via the AI"
                            + " engine, and stores the result in the StirlingPDFClassification"
-                            + " metadata field. Dispatched by the Classification policy; not"
-                            + " intended for direct client use.")
+                            + " metadata field. A document that already carries a verdict is"
+                            + " passed through untouched unless reclassify=true.")
    public ResponseEntity<Resource> classifyAndLabel(
-            @RequestParam("fileInput") MultipartFile fileInput) throws IOException {
+            @RequestParam("fileInput") MultipartFile fileInput,
+            @RequestParam(value = "reclassify", defaultValue = "false") boolean reclassify)
+            throws IOException {
        aiFeatureGate.requireClassify();
        try (PDDocument document = pdfDocumentFactory.load(fileInput, true)) {
            String fileName = safeFileName(fileInput.getOriginalFilename());

+            if (!reclassify && isClassified(document)) {
+                // Classifying twice costs a second engine call and charges for it, and a document
+                // that already carries a verdict has nothing new to learn. A pipeline can run this
+                // step over a mixed batch without paying for the ones already done.
+                log.debug("[classify-and-label] {} already classified; passing through", fileName);
+                return WebResponseUtils.pdfDocToWebResponse(document, fileName, tempFileManager);
+            }
+
            List<EngineLabel> allowed = resolveAllowedLabels();
            if (allowed.isEmpty()) {
                // No vocabulary to classify against: pass the file through unlabelled rather than
@@ -135,6 +151,25 @@ public class ClassifyLabelController {
        }
    }

+    /**
+     * Whether a verdict is already on the document.
+     *
+     * <p>This only reads back what a previous run of this step wrote. It is not a statement that
+     * the verdict is trustworthy: the key is ordinary PDF metadata that whoever supplied the file
+     * can set. Skipping the engine on the strength of it is safe because the cost of being wrong is
+     * a missing re-classification, not a wrong decision. Anything that makes a SECURITY decision
+     * from this field - routing a document somewhere on the strength of its label, say - must
+     * classify with {@code reclassify=true} rather than trust what arrived.
+     */
+    private static boolean isClassified(PDDocument document) {
+        PDDocumentInformation info = document.getDocumentInformation();
+        if (info == null) {
+            return false;
+        }
+        String existing = info.getCustomMetadataValue(PdfMetadataService.CLASSIFICATION_KEY);
+        return existing != null && !existing.isBlank();
+    }
+
    private List<AiPageText> extractWindow(PDDocument document) throws IOException {
        List<AiPageText> pages = new ArrayList<>();
        for (int pageNumber : windowPageNumbers(document.getNumberOfPages(), WINDOW_PAGES)) {
@@ -2,6 +2,7 @@ package stirling.software.proprietary.controller.api;

 import org.springframework.http.HttpStatus;
 import org.springframework.http.ResponseEntity;
+import org.springframework.security.access.prepost.PreAuthorize;
 import org.springframework.web.bind.annotation.GetMapping;
 import org.springframework.web.bind.annotation.RequestParam;

@@ -11,19 +12,25 @@ import lombok.RequiredArgsConstructor;

 import stirling.software.common.annotations.api.ProprietaryUiDataApi;
 import stirling.software.proprietary.audit.PortalAuditScope;
-import stirling.software.proprietary.audit.PortalAuditScopeResolver;
+import stirling.software.proprietary.audit.PortalDocumentsScopeResolver;
 import stirling.software.proprietary.model.api.documents.PortalDocumentsResponseDto;
-import stirling.software.proprietary.security.config.EnterpriseEndpoint;
 import stirling.software.proprietary.service.PortalDocumentsService;

-/** Serves the portal Documents review queue, derived from real audit data and scoped per caller. */
+/**
+ * Serves the portal Documents review queue, derived from real audit data and scoped per caller.
+ *
+ * <p>Open to every portal user (not Enterprise-gated): the Documents tab is a core Processor
+ * feature. Access is enforced by {@code @resourceAccess.canUsePortal()}; visibility is then
+ * resolved per deployment - self-hosted portal users see the whole server, SaaS users see their
+ * team (see {@link PortalDocumentsScopeResolver}).
+ */
@ProprietaryUiDataApi
@RequiredArgsConstructor
-@EnterpriseEndpoint
+@PreAuthorize("@resourceAccess.canUsePortal()")
 public class PortalDocumentsController {

    private final PortalDocumentsService portalDocumentsService;
-    private final PortalAuditScopeResolver auditScopeResolver;
+    private final PortalDocumentsScopeResolver documentsScopeResolver;

    // tier accepted for mock-seam symmetry; ignored (queue isn't tier-scoped).
    @GetMapping("/documents")
@@ -32,8 +39,9 @@ public class PortalDocumentsController {
            description = "Files processed through the org, derived from the audit trail.")
    public ResponseEntity<PortalDocumentsResponseDto> getDocuments(
            @RequestParam(value = "tier", required = false) String tier) {
-        PortalAuditScope scope = auditScopeResolver.resolve();
+        PortalAuditScope scope = documentsScopeResolver.resolve();
        if (!scope.allowed()) {
+            // SaaS caller with no team has nothing to show; surface an empty tab, not a 500.
            return ResponseEntity.status(HttpStatus.FORBIDDEN).build();
        }
        PortalDocumentsResponseDto body =
@@ -85,6 +85,11 @@ public record Policy(
        return new Policy(id, name, owner, enabled, inputs, steps, resolved, outputIds, teamId);
    }

+    /** A copy under a different owner (e.g. moving a seed off a placeholder name). */
+    public Policy withOwner(String newOwner) {
+        return new Policy(id, name, newOwner, enabled, inputs, steps, output, outputIds, teamId);
+    }
+
    /** A copy referencing the given saved output destinations. */
    public Policy withOutputIds(List<String> newOutputIds) {
        return new Policy(id, name, owner, enabled, inputs, steps, output, newOutputIds, teamId);
@@ -22,8 +22,8 @@ import stirling.software.proprietary.security.repository.TeamRepository;
 import stirling.software.proprietary.security.service.TeamService;

 /**
- * Seeds an enabled Classification policy per team so classification is on by default. Idempotent;
- * skips the internal team.
+ * Seeds an enabled Classification policy per team; idempotent, skips the internal team. Left
+ * unowned: nobody created it, and an owner here would have to name a real user.
 */
@Slf4j
@Component
@@ -34,6 +34,11 @@ public class DefaultClassificationPolicySeeder {
    private static final String CLASSIFY_ENDPOINT = "/api/v1/ai/tools/classify-and-label";
    private static final String POLICY_NAME = "Classification Policy";

+    /**
+     * Pre-existing seeds used this placeholder, which was never a user; see {@link #repairOwner}.
+     */
+    private static final String LEGACY_OWNER = "system";
+
    private final PolicyStore policyStore;
    private final TeamRepository teamRepository;

@@ -46,9 +51,8 @@ public class DefaultClassificationPolicySeeder {
                .ifPresent(team -> seedIfMissing(team.getId(), team.getName()));
    }

-    // Any team created at runtime (admin-created, SaaS sign-ups). Seeds inside the team's own
-    // transaction: rollback still leaves no policy behind, and the store's pessimistic lock needs a
-    // live transaction, which AFTER_COMMIT cannot offer.
+    // Seeds inside the new team's own transaction: rollback leaves no policy behind, and the
+    // store's pessimistic lock needs a live transaction, which AFTER_COMMIT cannot offer.
    @TransactionalEventListener(phase = TransactionPhase.BEFORE_COMMIT)
    public void onTeamCreated(TeamCreatedEvent event) {
        seedIfMissing(event.teamId(), event.teamName());
@@ -58,16 +62,33 @@ public class DefaultClassificationPolicySeeder {
        if (teamId == null || TeamService.INTERNAL_TEAM_NAME.equals(teamName)) {
            return;
        }
-        boolean alreadySeeded =
+        Policy existing =
                policyStore.findByTeam(teamId).stream()
-                        .anyMatch(DefaultClassificationPolicySeeder::isClassification);
-        if (alreadySeeded) {
+                        .filter(DefaultClassificationPolicySeeder::isClassification)
+                        .findFirst()
+                        .orElse(null);
+        if (existing != null) {
+            repairOwner(existing);
            return;
        }
        policyStore.save(defaultPolicy(teamId));
        log.info("Seeded default Classification policy for team {}", teamId);
    }

+    /**
+     * Clear an owner seeded as a placeholder name. An owner someone deliberately set is left alone.
+     */
+    private void repairOwner(Policy policy) {
+        if (!LEGACY_OWNER.equals(policy.owner())) {
+            return;
+        }
+        policyStore.save(policy.withOwner(null));
+        log.info(
+                "Cleared placeholder owner '{}' on Classification policy {}",
+                LEGACY_OWNER,
+                policy.id());
+    }
+
    private static boolean isClassification(Policy policy) {
        return policy.output() != null
                && CATEGORY.equals(policy.output().options().get("categoryId"));
@@ -85,7 +106,9 @@ public class DefaultClassificationPolicySeeder {
        return new Policy(
                null,
                POLICY_NAME,
-                "system",
+                // Nobody created this - it is seeded. A name here would have to be a real user, and
+                // every consumer of owner already handles its absence.
+                null,
                true,
                List.of(),
                List.of(new PipelineStep(CLASSIFY_ENDPOINT, Map.of())),
@@ -5,13 +5,13 @@ import java.time.temporal.ChronoUnit;
 import java.util.List;
 import java.util.concurrent.TimeUnit;

+import org.springframework.beans.factory.annotation.Qualifier;
 import org.springframework.data.domain.PageRequest;
 import org.springframework.data.domain.Sort;
 import org.springframework.scheduling.annotation.Scheduled;
 import org.springframework.stereotype.Service;
 import org.springframework.transaction.annotation.Transactional;

-import lombok.RequiredArgsConstructor;
 import lombok.extern.slf4j.Slf4j;

 import stirling.software.proprietary.config.AuditConfigurationProperties;
@@ -20,15 +20,31 @@ import stirling.software.proprietary.repository.PersistentAuditEventRepository;
 /** Service to periodically clean up old audit events based on retention policy. */
@Slf4j
@Service
-@RequiredArgsConstructor
 public class AuditCleanupService {

    private final PersistentAuditEventRepository auditRepository;
    private final AuditConfigurationProperties auditConfig;
+    private final boolean runningEE;

    // Default batch size for deletions
    private static final int BATCH_SIZE = 10000;

+    /**
+     * Maximum audit retention on non-Enterprise instances. Audit events feed the Documents tab on
+     * every instance, but longer history is an Enterprise feature - so non-EE deployments keep a
+     * shorter window ("infinite" included), bounding the always-on trail off-license.
+     */
+    private static final int NON_EE_MAX_RETENTION_DAYS = 30;
+
+    public AuditCleanupService(
+            PersistentAuditEventRepository auditRepository,
+            AuditConfigurationProperties auditConfig,
+            @Qualifier("runningEE") boolean runningEE) {
+        this.auditRepository = auditRepository;
+        this.auditConfig = auditConfig;
+        this.runningEE = runningEE;
+    }
+
    /**
     * Scheduled task that runs daily to clean up old audit events. The retention period is
     * configurable in settings.yml.
@@ -39,7 +55,7 @@ public class AuditCleanupService {
            return;
        }

-        int retentionDays = auditConfig.getRetentionDays();
+        int retentionDays = effectiveRetentionDays();
        if (retentionDays <= 0) {
            return;
        }
@@ -58,6 +74,20 @@ public class AuditCleanupService {
        }
    }

+    /**
+     * The retention window actually applied. Enterprise uses the configured value (0 = infinite);
+     * non-Enterprise is clamped to {@link #NON_EE_MAX_RETENTION_DAYS}.
+     */
+    int effectiveRetentionDays() {
+        int configured = auditConfig.getRetentionDays();
+        if (runningEE) {
+            return configured;
+        }
+        return configured <= 0
+                ? NON_EE_MAX_RETENTION_DAYS
+                : Math.min(configured, NON_EE_MAX_RETENTION_DAYS);
+    }
+
    /**
     * Performs batch deletion of events to prevent long-running transactions and potential database
     * locks.
@@ -87,10 +87,7 @@ public class AuditService {
     * @param level The minimum audit level required for this event to be logged
     */
    public void audit(AuditEventType type, Map<String, Object> data, AuditLevel level) {
-        // Skip auditing if this level is not enabled or if not Enterprise edition
-        if (!auditConfig.isEnabled()
-                || !auditConfig.getAuditLevel().includes(level)
-                || !runningEE) {
+        if (!shouldRecord(type, level)) {
            return;
        }

@@ -126,8 +123,7 @@ public class AuditService {
     */
    public void audit(
            String principal, AuditEventType type, Map<String, Object> data, AuditLevel level) {
-        // Skip auditing if this level is not enabled or if not Enterprise edition
-        if (!auditConfig.isLevelEnabled(level) || !runningEE) {
+        if (!shouldRecord(type, level)) {
            return;
        }

@@ -156,8 +152,7 @@ public class AuditService {
     * @param level The minimum audit level required for this event to be logged
     */
    public void audit(String type, Map<String, Object> data, AuditLevel level) {
-        // Skip auditing if this level is not enabled or if not Enterprise edition
-        if (!auditConfig.isLevelEnabled(level) || !runningEE) {
+        if (!shouldRecord(type, level)) {
            return;
        }

@@ -192,8 +187,7 @@ public class AuditService {
     * @param level The minimum audit level required for this event to be logged
     */
    public void audit(String principal, String type, Map<String, Object> data, AuditLevel level) {
-        // Skip auditing if this level is not enabled or if not Enterprise edition
-        if (!auditConfig.isLevelEnabled(level) || !runningEE) {
+        if (!shouldRecord(type, level)) {
            return;
        }

@@ -223,9 +217,7 @@ public class AuditService {
            AuditEventType type,
            Map<String, Object> data,
            AuditLevel level) {
-        if (!auditConfig.isEnabled()
-                || !auditConfig.getAuditLevel().includes(level)
-                || !runningEE) {
+        if (!shouldRecord(type, level)) {
            return;
        }

@@ -250,9 +242,7 @@ public class AuditService {
            String type,
            Map<String, Object> data,
            AuditLevel level) {
-        if (!auditConfig.isEnabled()
-                || !auditConfig.getAuditLevel().includes(level)
-                || !runningEE) {
+        if (!shouldRecord(type, level)) {
            return;
        }

@@ -626,6 +616,55 @@ public class AuditService {
        return auditConfig.getAuditLevel().includes(requiredLevel);
    }

+    /**
+     * Type-aware variant used by the controller aspect, which resolves the event type before
+     * deciding whether to record. Document-processing events feed the Documents tab (available to
+     * every Processor user), so they audit without an Enterprise license; the rest of the audit log
+     * stays Enterprise-only.
+     */
+    public boolean shouldAudit(
+            AuditEventType eventType, Method method, AuditConfigurationProperties auditConfig) {
+        if (!auditConfig.isEnabled() || !isLicensedToRecord(eventType)) {
+            return false;
+        }
+
+        Audited auditedAnnotation = method.getAnnotation(Audited.class);
+        AuditLevel requiredLevel =
+                (auditedAnnotation != null) ? auditedAnnotation.level() : AuditLevel.BASIC;
+
+        return auditConfig.getAuditLevel().includes(requiredLevel);
+    }
+
+    /**
+     * Whether an event of this type and level should be persisted: the configured audit level must
+     * include it and the current license must permit recording it.
+     */
+    private boolean shouldRecord(AuditEventType type, AuditLevel level) {
+        return auditConfig.isLevelEnabled(level) && isLicensedToRecord(type);
+    }
+
+    private boolean shouldRecord(String type, AuditLevel level) {
+        return auditConfig.isLevelEnabled(level) && isLicensedToRecord(type);
+    }
+
+    /**
+     * Whether the current license permits recording this event type. Enterprise records everything;
+     * without it only document-processing events (PDF_PROCESS, FILE_OPERATION) are captured,
+     * because they back the Documents tab that is open to every Processor user (still subject to
+     * audit being enabled at a level that includes them). Everything else stays Enterprise-only.
+     */
+    private boolean isLicensedToRecord(AuditEventType type) {
+        return runningEE
+                || type == AuditEventType.PDF_PROCESS
+                || type == AuditEventType.FILE_OPERATION;
+    }
+
+    private boolean isLicensedToRecord(String type) {
+        return runningEE
+                || AuditEventType.PDF_PROCESS.name().equals(type)
+                || AuditEventType.FILE_OPERATION.name().equals(type);
+    }
+
    /**
     * Add timing and response status data to the audit record
     *
@@ -76,7 +76,8 @@ class ControllerAuditAspectTest {
        @DisplayName("shouldAudit false proceeds without recording")
        void skipsWhenShouldAuditFalse() throws Throwable {
            ProceedingJoinPoint jp = joinPointFor("getEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(false);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(false);
            when(jp.proceed()).thenReturn("ok");

            Object result = aspect.auditGetMethod(jp);
@@ -102,7 +103,8 @@ class ControllerAuditAspectTest {
        @DisplayName("records success outcome and returns result")
        void recordsSuccess() throws Throwable {
            ProceedingJoinPoint jp = joinPointFor("postEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(true);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(true);
            when(auditService.captureCurrentPrincipal()).thenReturn("alice");
            when(auditService.captureCurrentOrigin()).thenReturn("WEB");
            when(auditService.createBaseAuditData(eq(jp), any(AuditLevel.class)))
@@ -134,7 +136,8 @@ class ControllerAuditAspectTest {
            MDC.put("auditPrincipal", "fromMdc");
            MDC.put("auditOrigin", "API");
            ProceedingJoinPoint jp = joinPointFor("postEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(true);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(true);
            when(auditService.createBaseAuditData(eq(jp), any(AuditLevel.class)))
                    .thenReturn(new HashMap<>());
            when(auditService.resolveEventType(
@@ -166,7 +169,8 @@ class ControllerAuditAspectTest {
        @DisplayName("records failure outcome and rethrows")
        void recordsFailureAndRethrows() throws Throwable {
            ProceedingJoinPoint jp = joinPointFor("postEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(true);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(true);
            when(auditService.captureCurrentPrincipal()).thenReturn("alice");
            when(auditService.captureCurrentOrigin()).thenReturn("WEB");
            when(auditService.createBaseAuditData(eq(jp), any(AuditLevel.class)))
@@ -204,7 +208,8 @@ class ControllerAuditAspectTest {
        @DisplayName("annotated method proceeds without double-auditing")
        void annotatedMethodSkips() throws Throwable {
            ProceedingJoinPoint jp = joinPointFor("annotatedEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(true);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(true);
            when(auditService.captureCurrentPrincipal()).thenReturn("alice");
            when(auditService.captureCurrentOrigin()).thenReturn("WEB");
            when(jp.proceed()).thenReturn("ok");
@@ -232,7 +237,8 @@ class ControllerAuditAspectTest {
        @DisplayName("captures result when enabled and non-UI type")
        void capturesResult() throws Throwable {
            ProceedingJoinPoint jp = joinPointFor("postEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(true);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(true);
            when(auditService.captureCurrentPrincipal()).thenReturn("alice");
            when(auditService.captureCurrentOrigin()).thenReturn("WEB");
            when(auditService.createBaseAuditData(eq(jp), any(AuditLevel.class)))
@@ -262,7 +268,8 @@ class ControllerAuditAspectTest {
        @DisplayName("UI_DATA result is not captured")
        void uiDataResultSkipped() throws Throwable {
            ProceedingJoinPoint jp = joinPointFor("getEndpoint");
-            when(auditService.shouldAudit(any(Method.class), eq(auditConfig))).thenReturn(true);
+            when(auditService.shouldAudit(any(), any(Method.class), eq(auditConfig)))
+                    .thenReturn(true);
            when(auditService.captureCurrentPrincipal()).thenReturn("alice");
            when(auditService.captureCurrentOrigin()).thenReturn("WEB");
            when(auditService.createBaseAuditData(eq(jp), any(AuditLevel.class)))
@@ -14,6 +14,7 @@ import static org.mockito.Mockito.when;
 import java.util.List;

 import org.apache.pdfbox.pdmodel.PDDocument;
+import org.apache.pdfbox.pdmodel.PDDocumentInformation;
 import org.junit.jupiter.api.Test;
 import org.junit.jupiter.api.extension.ExtendWith;
 import org.mockito.ArgumentCaptor;
@@ -76,7 +77,7 @@ class ClassifyLabelControllerTest {
                .thenReturn("{\"outcome\":\"classification\",\"labels\":[\"invoice\"]}");

        try {
-            controller.classifyAndLabel(file);
+            controller.classifyAndLabel(file, false);
        } catch (Exception ignored) {
            // WebResponseUtils.pdfDocToWebResponse needs a real temp file; the engine call and
            // metadata write we assert on have already happened by the time it runs.
@@ -89,6 +90,52 @@ class ClassifyLabelControllerTest {
        return objectMapper.readTree(body.getValue());
    }

+    /** Stubs a document that already carries a verdict, as a second run over a batch would see. */
+    private MultipartFile alreadyClassifiedDocument() throws Exception {
+        PDDocument document = mock(PDDocument.class);
+        PDDocumentInformation info = mock(PDDocumentInformation.class);
+        when(document.getDocumentInformation()).thenReturn(info);
+        when(info.getCustomMetadataValue(PdfMetadataService.CLASSIFICATION_KEY))
+                .thenReturn("{\"labels\":[\"invoice\"]}");
+        MultipartFile file = mock(MultipartFile.class);
+        when(file.getOriginalFilename()).thenReturn("invoice.pdf");
+        when(pdfDocumentFactory.load(any(MultipartFile.class), eq(true))).thenReturn(document);
+        return file;
+    }
+
+    @Test
+    void classifyAndLabel_skipsADocumentThatAlreadyCarriesAVerdict() throws Exception {
+        withLabels(List.of(new ClassificationLabel("invoice", "Invoice", null)));
+        MultipartFile file = alreadyClassifiedDocument();
+
+        try {
+            controller.classifyAndLabel(file, false);
+        } catch (Exception ignored) {
+            // The response needs a real temp file; the decision under test happens before it.
+        }
+
+        // No second engine call, and no charge for one: re-classifying buys the same answer twice.
+        verify(aiEngineClient, never()).post(anyString(), anyString(), any());
+        verify(pdfMetadataService, never()).setClassificationMetadata(any(), anyString());
+    }
+
+    @Test
+    void classifyAndLabel_reclassifiesWhenAskedTo() throws Exception {
+        withLabels(List.of(new ClassificationLabel("invoice", "Invoice", null)));
+        MultipartFile file = alreadyClassifiedDocument();
+        when(pdfContentExtractor.extractPageTextRaw(any(), eq(1))).thenReturn("Invoice total");
+        when(aiEngineClient.post(eq("/api/v1/documents/classify"), anyString(), isNull()))
+                .thenReturn("{\"outcome\":\"classification\",\"labels\":[\"receipt\"]}");
+
+        try {
+            controller.classifyAndLabel(file, true);
+        } catch (Exception ignored) {
+            // As above.
+        }
+
+        verify(aiEngineClient).post(eq("/api/v1/documents/classify"), anyString(), isNull());
+    }
+
    @Test
    void classifyAndLabel_writesClassificationWithoutOutcome() throws Exception {
        withLabels(List.of(new ClassificationLabel("invoice", "Invoice", null)));
@@ -36,10 +36,14 @@ class DefaultClassificationPolicySeederTest {
    }

    private static Policy classificationPolicy(Long teamId) {
+        return classificationPolicy(teamId, null);
+    }
+
+    private static Policy classificationPolicy(Long teamId, String owner) {
        return new Policy(
                "p1",
                "Classification Policy",
-                "system",
+                owner,
                true,
                List.of(),
                List.of(),
@@ -77,6 +81,29 @@ class DefaultClassificationPolicySeederTest {
        verify(policyStore, never()).save(any());
    }

+    @Test
+    void clearsAPlaceholderOwnerSeededBeforeOwnersHadToBeReal() {
+        when(policyStore.findByTeam(7L)).thenReturn(List.of(classificationPolicy(7L, "system")));
+
+        seeder().onTeamCreated(new TeamCreatedEvent(7L, "Acme"));
+
+        // "system" was never a user row, and a step dispatch authenticates as the owner. Absence
+        // is handled everywhere; a placeholder name is not.
+        ArgumentCaptor<Policy> saved = ArgumentCaptor.forClass(Policy.class);
+        verify(policyStore).save(saved.capture());
+        assertThat(saved.getValue().owner()).isNull();
+        assertThat(saved.getValue().id()).isEqualTo("p1");
+    }
+
+    @Test
+    void leavesADeliberatelyChosenOwnerAlone() {
+        when(policyStore.findByTeam(7L)).thenReturn(List.of(classificationPolicy(7L, "alice")));
+
+        seeder().onTeamCreated(new TeamCreatedEvent(7L, "Acme"));
+
+        verify(policyStore, never()).save(any());
+    }
+
    @Test
    void doesNotSeedForTheInternalTeam() {
        seeder().onTeamCreated(new TeamCreatedEvent(2L, "Internal"));
@@ -0,0 +1,55 @@
+package stirling.software.proprietary.service;
+
+import static org.assertj.core.api.Assertions.assertThat;
+import static org.mockito.Mockito.mock;
+
+import org.junit.jupiter.api.DisplayName;
+import org.junit.jupiter.api.Test;
+
+import stirling.software.common.model.ApplicationProperties;
+import stirling.software.proprietary.config.AuditConfigurationProperties;
+import stirling.software.proprietary.repository.PersistentAuditEventRepository;
+
+class AuditCleanupServiceTest {
+
+    private final PersistentAuditEventRepository repository =
+            mock(PersistentAuditEventRepository.class);
+
+    private AuditCleanupService service(boolean runningEE, int retentionDays) {
+        ApplicationProperties props = new ApplicationProperties();
+        var audit = props.getPremium().getEnterpriseFeatures().getAudit();
+        audit.setEnabled(true);
+        audit.setRetentionDays(retentionDays);
+        return new AuditCleanupService(
+                repository, new AuditConfigurationProperties(props), runningEE);
+    }
+
+    @Test
+    @DisplayName("Enterprise keeps the configured retention, including infinite")
+    void enterpriseUsesConfigured() {
+        assertThat(service(true, 90).effectiveRetentionDays()).isEqualTo(90);
+        assertThat(service(true, 365).effectiveRetentionDays()).isEqualTo(365);
+        assertThat(service(true, 0).effectiveRetentionDays()).isEqualTo(0);
+    }
+
+    @Test
+    @DisplayName("non-Enterprise caps retention at 30 days")
+    void nonEnterpriseCapsHigherValues() {
+        assertThat(service(false, 90).effectiveRetentionDays()).isEqualTo(30);
+        assertThat(service(false, 365).effectiveRetentionDays()).isEqualTo(30);
+    }
+
+    @Test
+    @DisplayName("non-Enterprise respects a shorter configured retention")
+    void nonEnterpriseRespectsLower() {
+        assertThat(service(false, 14).effectiveRetentionDays()).isEqualTo(14);
+        assertThat(service(false, 7).effectiveRetentionDays()).isEqualTo(7);
+    }
+
+    @Test
+    @DisplayName("non-Enterprise cannot retain forever (<= 0 becomes the cap)")
+    void nonEnterpriseNoInfinite() {
+        assertThat(service(false, 0).effectiveRetentionDays()).isEqualTo(30);
+        assertThat(service(false, -1).effectiveRetentionDays()).isEqualTo(30);
+    }
+}
@@ -166,6 +166,42 @@ class AuditServiceTest {
            verify(repository, never()).add(any(AuditEvent.class));
        }

+        @Test
+        @DisplayName("records document-processing events even without EE (Documents feed)")
+        void recordsDocumentEventsWithoutEE() {
+            AuditService nonEe =
+                    new AuditService(
+                            repository, auditConfig, false, pdfDocumentFactory, jwtService);
+            authenticateAs("alice");
+
+            nonEe.audit(AuditEventType.PDF_PROCESS, new HashMap<>(), AuditLevel.BASIC);
+            nonEe.audit(AuditEventType.FILE_OPERATION, new HashMap<>(), AuditLevel.BASIC);
+
+            org.mockito.ArgumentCaptor<AuditEvent> captor =
+                    org.mockito.ArgumentCaptor.forClass(AuditEvent.class);
+            verify(repository, org.mockito.Mockito.times(2)).add(captor.capture());
+            assertThat(captor.getAllValues())
+                    .extracting(AuditEvent::getType)
+                    .containsExactly(
+                            AuditEventType.PDF_PROCESS.name(),
+                            AuditEventType.FILE_OPERATION.name());
+        }
+
+        @Test
+        @DisplayName("type-aware shouldAudit lets doc events through without EE, blocks others")
+        void typeAwareShouldAuditWithoutEE() throws Exception {
+            AuditService nonEe =
+                    new AuditService(
+                            repository, auditConfig, false, pdfDocumentFactory, jwtService);
+            Method m = Object.class.getMethod("toString");
+
+            assertThat(nonEe.shouldAudit(AuditEventType.PDF_PROCESS, m, auditConfig)).isTrue();
+            assertThat(nonEe.shouldAudit(AuditEventType.FILE_OPERATION, m, auditConfig)).isTrue();
+            assertThat(nonEe.shouldAudit(AuditEventType.USER_LOGIN, m, auditConfig)).isFalse();
+            // With EE, non-doc events at/under the configured level audit too.
+            assertThat(service.shouldAudit(AuditEventType.USER_LOGIN, m, auditConfig)).isTrue();
+        }
+
        @Test
        @DisplayName("skips when audit disabled")
        void skipsWhenDisabled() {
@@ -0,0 +1,46 @@
+package stirling.software.saas.security;
+
+import java.util.List;
+import java.util.Objects;
+
+import org.springframework.context.annotation.Primary;
+import org.springframework.context.annotation.Profile;
+import org.springframework.stereotype.Component;
+
+import lombok.RequiredArgsConstructor;
+
+import stirling.software.proprietary.audit.PortalAuditScope;
+import stirling.software.proprietary.audit.PortalAuditScopeResolver;
+import stirling.software.proprietary.audit.PortalDocumentsScopeResolver;
+import stirling.software.proprietary.security.repository.TeamMembershipRepository;
+
+/**
+ * SaaS documents visibility: platform admins see the whole server; every other portal user sees
+ * their own team's documents (by member email).
+ */
+@Component
+@Primary
+@Profile("saas")
+@RequiredArgsConstructor
+public class SaasPortalDocumentsScopeResolver implements PortalDocumentsScopeResolver {
+
+    private final TeamSecurityExpressions teamSecurity;
+    private final TeamMembershipRepository membershipRepository;
+
+    @Override
+    public PortalAuditScope resolve() {
+        if (PortalAuditScopeResolver.hasAdminAuthority()) {
+            return PortalAuditScope.server();
+        }
+        Long teamId = teamSecurity.currentUserTeamId();
+        if (teamId == null) {
+            return PortalAuditScope.denied();
+        }
+        List<String> memberEmails =
+                membershipRepository.findByTeamId(teamId).stream()
+                        .map(m -> m.getUser() == null ? null : m.getUser().getEmail())
+                        .filter(Objects::nonNull)
+                        .toList();
+        return PortalAuditScope.team("team:" + teamId, memberEmails);
+    }
+}
@@ -67,6 +67,30 @@ COPY --from=app-build /app/app/core/build/libs/*.jar app.jar
 RUN java -Djarmode=tools -jar app.jar extract --layers --destination /layers


+# Stage 2b: AI engine. Built at its final path so the venv resolves after the copy, on uv's
+# managed CPython because the runtime base ships Python 3.12.
+FROM ghcr.io/astral-sh/uv:bookworm-slim@sha256:22334efe746f1b69217d455049b484d7b8cacfb2d5f42555580b62415a98e0a3 AS engine-build
+ENV UV_PYTHON_INSTALL_DIR=/opt/stirling-engine/python
+WORKDIR /opt/stirling-engine
+COPY engine/pyproject.toml engine/uv.lock ./
+# One layer: trimming in a second RUN would cache the untrimmed copy too, and this build
+# exports every layer to a GHA cache that is capped repo-wide. Trimming saves ~20MB.
+RUN --mount=type=cache,target=/root/.cache/uv \
+    set -eux; \
+    apt-get update && apt-get install -y --no-install-recommends binutils; \
+    uv python install 3.13; \
+    uv sync --frozen --no-dev --no-install-project --group engine --python-preference only-managed; \
+    P="$(ls -d /opt/stirling-engine/python/cpython-*)"; \
+    rm -rf "$P/share" "$P/include" \
+           "$P/lib/python3.13/idlelib" "$P/lib/python3.13/tkinter" \
+           "$P/lib/python3.13/ensurepip" "$P/lib/python3.13/pydoc_data" \
+           "$P/lib/python3.13/test" "$P/lib/python3.13/lib2to3"; \
+    find /opt/stirling-engine -name '__pycache__' -type d -prune -exec rm -rf {} + ; \
+    find /opt/stirling-engine \( -name '*.so' -o -name '*.so.*' \) -print0 \
+      | xargs -0 -r strip --strip-unneeded 2>/dev/null || true; \
+    apt-get purge -y binutils; apt-get autoremove -y; rm -rf /var/lib/apt/lists/*
+
+
 # Stage 3: Final runtime image on top of pre-built base
 FROM ${BASE_IMAGE}

@@ -84,6 +108,12 @@ COPY --link --from=app-build --chown=1000:1000 \
     /app/build/libs/restart-helper.jar /restart-helper.jar
 COPY --link --chown=1000:1000 scripts/ /scripts/

+# init-without-ocr.sh starts the engine when this directory exists, so other images are unaffected.
+COPY --link --from=engine-build --chown=1000:1000 /opt/stirling-engine/python /opt/stirling-engine/python
+COPY --link --from=engine-build --chown=1000:1000 /opt/stirling-engine/.venv  /opt/stirling-engine/.venv
+COPY --link --chown=1000:1000 engine/.env  /opt/stirling-engine/.env
+COPY --link --chown=1000:1000 engine/src/  /opt/stirling-engine/src/
+
 # Fonts go to system dir, root ownership is correct (world-readable)
 COPY app/core/src/main/resources/static/fonts/*.ttf /usr/share/fonts/truetype/

@@ -97,6 +127,8 @@ RUN set -eux; \
    ln -s /storage /app/storage; \
    chown -h stirlingpdfuser:stirlingpdfgroup /app/logs /app/configs /app/customFiles /app/pipeline /app/storage; \
    chown stirlingpdfuser:stirlingpdfgroup /app; \
+    mkdir -p /opt/stirling-engine/data; \
+    chown -R stirlingpdfuser:stirlingpdfgroup /opt/stirling-engine/data; \
    chmod 750 /tmp/stirling-pdf; \
    chmod 750 /tmp/stirling-pdf/heap_dumps; \
    fc-cache -f
@@ -116,6 +148,10 @@ ENV VERSION_TAG=$VERSION_TAG \
    PGID=1000 \
    UMASK=022 \
    FAT_DOCKER=true \
+    AIENGINE_ENABLED=true \
+    STIRLING_ENGINE_HOME=/opt/stirling-engine \
+    STIRLING_ENGINE_PORT=5001 \
+    STIRLING_ENGINE_WORKERS=2 \
    INSTALL_BOOK_AND_ADVANCED_HTML_OPS=false \
    STIRLING_TEMPFILES_DIRECTORY=/tmp/stirling-pdf \
    TMPDIR=/tmp/stirling-pdf \
@@ -1,35 +1,43 @@
 # syntax=docker/dockerfile:1.5
-FROM ghcr.io/astral-sh/uv:python3.13-bookworm-slim@sha256:531f855bda2c73cd6ef67d56b733b357cea384185b3022bd09f05e002cd144ca

-ARG TASK_VERSION=3.52.0
-RUN apt-get update \
-    && apt-get install -y --no-install-recommends curl ca-certificates \
-    && ARCH=$(dpkg --print-architecture) \
-    && curl -fsSL "https://github.com/go-task/task/releases/download/v${TASK_VERSION}/task_${TASK_VERSION}_linux_${ARCH}.deb" -o /tmp/task.deb \
-    && dpkg -i /tmp/task.deb \
-    && rm /tmp/task.deb \
-    && rm -rf /var/lib/apt/lists/*
+# uv resolves the venv here so its ~52MB binary stays out of the runtime image.
+FROM ghcr.io/astral-sh/uv:python3.13-bookworm-slim@sha256:531f855bda2c73cd6ef67d56b733b357cea384185b3022bd09f05e002cd144ca AS builder

-# Source under /app/engine/ to match root Taskfile's `includes.engine.dir: engine`.
 WORKDIR /app/engine
-COPY engine/pyproject.toml engine/uv.lock engine/.env ./
-COPY engine/scripts/ ./scripts/
+COPY engine/pyproject.toml engine/uv.lock ./
 RUN --mount=type=cache,target=/root/.cache/uv \
-    uv sync --frozen --no-dev --group engine
+    uv sync --frozen --no-dev --no-install-project --group engine

-COPY engine/src/ ./src/
-WORKDIR /app
-COPY Taskfile.yml ./
-COPY .taskfiles/ ./.taskfiles/
+FROM python:3.13-slim-bookworm@sha256:00faa2debb87529f9f0764e9491d8ba400a3678976616c3bd7cb193745ac20d1 AS runtime
+
+# Created before the COPYs so they land owned; a later chown -R duplicates the venv layer.
+RUN set -eux; \
+    groupadd --system --gid 1000 stirling; \
+    useradd --system --uid 1000 --gid 1000 --home /app/engine stirling; \
+    mkdir -p /app/engine/data; \
+    chown stirling:stirling /app/engine /app/engine/data
+
+WORKDIR /app/engine
+COPY --from=builder --chown=stirling:stirling /app/engine/.venv ./.venv
+# settings.py resolves ENGINE_ROOT to /app/engine, so .env must sit here.
+COPY --chown=stirling:stirling engine/.env ./
+COPY --chown=stirling:stirling engine/src/ ./src/

 ENV PATH="/app/engine/.venv/bin:$PATH"
 ENV PYTHONUNBUFFERED=1
 ENV STIRLING_ENGINE_WORKERS=4
-# Container runs on a fixed port; skip the host-only free-port probe (its script
-# is not shipped in the image). engine:run honours these.
-ENV ENGINE_PORT_PROBE=false
 ENV STIRLING_ENGINE_PORT=5001
+# Fail closed: without a secret the document routes trust caller-supplied X-User-Id.
+# Set STIRLING_ENGINE_SHARED_SECRET (the backend sends it as X-Engine-Auth), or false to opt out.
+ENV STIRLING_ENGINE_REQUIRE_AUTH=true
+
+# `stirling` resolves from the working directory.
+WORKDIR /app/engine/src
+USER stirling

 EXPOSE 5001

-CMD ["task", "engine:run"]
+HEALTHCHECK --interval=30s --timeout=5s --start-period=40s --retries=3 \
+  CMD ["python", "-c", "import os,sys,urllib.request; sys.exit(0 if urllib.request.urlopen('http://127.0.0.1:%s/health' % os.environ.get('STIRLING_ENGINE_PORT','5001'), timeout=4).status==200 else 1)"]
+
+CMD ["sh", "-c", "exec uvicorn stirling.api.app:app --host 0.0.0.0 --port ${STIRLING_ENGINE_PORT:-5001} --workers ${STIRLING_ENGINE_WORKERS:-4}"]
@@ -16,8 +16,9 @@ engine = [
    "psycopg[binary,pool]>=3.3.4",
    "pydantic>=2.13.4",
    # <2 cap: 1.99.0 patches CVE-2026-46678; 2.0 is an untested major migration.
-    "pydantic-ai>=1.107.2,<2.0.0",
-    "pydantic-ai-slim[voyageai]>=1.107.2,<2.0.0",
+    # Explicit extras: the `pydantic-ai` meta-package pulls all 20 providers (~230MB).
+    # No `voyageai` extra either; stirling.documents.voyage speaks its API directly.
+    "pydantic-ai-slim[anthropic,openai]>=1.107.2,<2.0.0",
    "pydantic-settings>=2.15.0",
    "python-dotenv>=1.2.2",
    "sqlite-vec>=0.1.9",
@@ -70,6 +70,19 @@ Provider credentials (and any local overrides) go in the uncommitted
 VOYAGE_API_KEY=your-key
 ```

+### Embedding providers
+
+`STIRLING_RAG_EMBEDDING_MODEL` is a `provider:model` string. Any OpenAI-compatible
+`/v1/embeddings` endpoint (vLLM, Ollama, TEI, llama.cpp) works by pointing a base URL
+at it. Note `OPENAI_BASE_URL` is global and also redirects chat completions; push
+`provider`/`api_key`/`base_url` through admin AI settings to move embeddings only.
+Ollama reads `OLLAMA_BASE_URL`, and omitting it fails the first embed call, not startup.
+
+```
+STIRLING_RAG_EMBEDDING_MODEL=ollama:nomic-embed-text
+OLLAMA_BASE_URL=http://ollama:11434/v1
+```
+
 ## Backends

 **`sqlite`** - Embedded sqlite-vec. Single `.db` file, zero ops. Ideal for dev
@@ -6,6 +6,7 @@ from pydantic_ai.providers.openai import OpenAIProvider

 from stirling.documents.chunker import chunk_text
 from stirling.documents.store import Document
+from stirling.documents.voyage import build_voyage_model

 # Keep each upstream embed request under every major provider's per-call limit while
 # still batching large enough that a book-sized document ingests in a reasonable number
@@ -14,6 +15,9 @@ from stirling.documents.store import Document
 DEFAULT_EMBED_BATCH_SIZE = 256


+VOYAGE_PROVIDER = "voyageai"
+
+
 def _build_embedder(
    model_name: str,
    *,
@@ -23,11 +27,17 @@ def _build_embedder(
 ) -> Embedder:
    """Construct an :class:`Embedder`; explicit provider/api_key/base_url is the config-push path, else env form."""
    if not provider and not api_key and not base_url:
+        # Env form is a "provider:model" string; Voyage needs the SDK-free adapter.
+        env_provider, sep, env_model = model_name.partition(":")
+        if sep and env_provider.lower() == VOYAGE_PROVIDER:
+            return Embedder(build_voyage_model(env_model))
        return Embedder(model_name)

    provider_name = (provider or "").lower()
    key = api_key or None
-    if provider_name in ("voyageai", "openai"):
+    if provider_name == VOYAGE_PROVIDER:
+        return Embedder(build_voyage_model(model_name, api_key=key, base_url=base_url or None))
+    if provider_name == "openai":
        return Embedder(f"{provider_name}:{model_name}")
    if provider_name in ("ollama", "custom"):
        openai_provider = OpenAIProvider(base_url=base_url or None, api_key=key or "ollama")
@@ -2,9 +2,11 @@ from __future__ import annotations

 import asyncio
 import json
+import logging
 import math
 import re
 import sqlite3
+import time
 from datetime import UTC, datetime
 from pathlib import Path

@@ -19,11 +21,31 @@ _READ_PERMISSION = "read"
 # write lock. With multiple worker processes opening the same file, they collide on
 # startup schema-init and get "database is locked". Wait for the lock instead.
 _BUSY_TIMEOUT_MS = 5000
+# journal_mode answers SQLITE_BUSY without consulting the busy handler, so it needs its own retry.
+_WAL_SWITCH_ATTEMPTS = 10
+_WAL_RETRY_DELAY_S = 0.1
 # sqlite stores TIMESTAMP as TEXT. We normalise to UTC ISO 8601 ``YYYY-MM-DD HH:MM:SS``
 # so lexicographic comparison against ``datetime('now')`` matches chronological order.
 _SQLITE_DATETIME_FMT = "%Y-%m-%d %H:%M:%S"


+logger = logging.getLogger(__name__)
+
+
+def _enable_wal(conn: sqlite3.Connection) -> None:
+    """Switch the connection to WAL, tolerating workers racing to do the same."""
+    for _ in range(_WAL_SWITCH_ATTEMPTS):
+        try:
+            conn.execute("PRAGMA journal_mode=WAL")
+            return
+        except sqlite3.OperationalError:
+            row = conn.execute("PRAGMA journal_mode").fetchone()
+            if row is not None and str(row[0]).lower() == "wal":
+                return  # another worker won the race and already switched it
+            time.sleep(_WAL_RETRY_DELAY_S)
+    logger.warning("Could not switch the document store to WAL; continuing on the default journal mode.")
+
+
 def _to_sqlite_utc(dt: datetime | None) -> str | None:
    if dt is None:
        return None
@@ -58,7 +80,7 @@ class SqliteVecStore(DocumentStore):
        if self._db_path is not None:
            # Set before the WAL switch below: that pragma also takes the lock.
            conn.execute(f"PRAGMA busy_timeout={_BUSY_TIMEOUT_MS}")
-            conn.execute("PRAGMA journal_mode=WAL")
+            _enable_wal(conn)

        self._conn = conn
        self._lock = asyncio.Lock()
@@ -0,0 +1,57 @@
+"""VoyageAI embeddings over its OpenAI-shaped REST API.
+
+The `voyageai` SDK pulls PIL, numpy, tokenizers and langchain at import for multimodal,
+chunking and local-inference features the engine never uses (~207MB).
+"""
+
+from __future__ import annotations
+
+import os
+from collections.abc import Sequence
+
+from pydantic_ai.embeddings import EmbeddingResult, EmbeddingSettings
+from pydantic_ai.embeddings.openai import OpenAIEmbeddingModel
+from pydantic_ai.embeddings.result import EmbedInputType
+from pydantic_ai.providers.openai import OpenAIProvider
+
+VOYAGE_BASE_URL = "https://api.voyageai.com/v1"
+VOYAGE_API_KEY_ENV = "VOYAGE_API_KEY"
+
+# Keeps a keyless engine bootable, and stops the client falling back to OPENAI_API_KEY.
+_MISSING_API_KEY = "stirling-voyage-api-key-not-configured"
+
+
+class VoyageEmbeddingModel(OpenAIEmbeddingModel):
+    """Voyage embeddings spoken over the OpenAI wire format."""
+
+    async def embed(
+        self,
+        inputs: str | Sequence[str],
+        *,
+        input_type: EmbedInputType,
+        settings: EmbeddingSettings | None = None,
+    ) -> EmbeddingResult:
+        """Embed `inputs`, forwarding Voyage's `input_type` that the OpenAI model drops."""
+        if self._client.api_key == _MISSING_API_KEY:
+            raise ValueError(
+                f"VoyageAI embeddings need an API key: set {VOYAGE_API_KEY_ENV} or push one via admin AI settings."
+            )
+        merged: EmbeddingSettings = {**(settings or {})}
+        # extra_body is declared `object`, so narrow rather than assume a mapping.
+        current = merged.get("extra_body")
+        extra_body: dict[str, object] = dict(current) if isinstance(current, dict) else {}
+        extra_body.setdefault("input_type", input_type)
+        merged["extra_body"] = extra_body
+        return await super().embed(inputs, input_type=input_type, settings=merged)
+
+
+def build_voyage_model(
+    model_name: str,
+    *,
+    api_key: str | None = None,
+    base_url: str | None = None,
+) -> VoyageEmbeddingModel:
+    """Build a Voyage embedding model; a missing key only fails once an embed is attempted."""
+    key = api_key or os.environ.get(VOYAGE_API_KEY_ENV) or _MISSING_API_KEY
+    provider = OpenAIProvider(base_url=base_url or VOYAGE_BASE_URL, api_key=key)
+    return VoyageEmbeddingModel(model_name, provider=provider)
@@ -1,5 +1,7 @@
 from __future__ import annotations

+from pathlib import Path
+
 import pytest

 from stirling.contracts import PageText
@@ -647,3 +649,19 @@ def _dummy_tool_def() -> object:
    """Sentinel passed to ``_prepare_search_knowledge``. The callback only inspects
    ``_search_count``; it doesn't read anything off the tool_def or context."""
    return object()
+
+
+# concurrent store startup
+
+
+def test_many_stores_open_the_same_file_without_locking_out(tmp_path: Path) -> None:
+    """Workers all construct a store against one file on boot; the WAL switch races."""
+    import concurrent.futures
+
+    db_path = tmp_path / "rag.db"
+    with concurrent.futures.ThreadPoolExecutor(max_workers=8) as pool:
+        stores = list(pool.map(lambda _: SqliteVecStore(db_path), range(8)))
+
+    assert len(stores) == 8
+    mode = stores[0]._conn.execute("PRAGMA journal_mode").fetchone()[0]
+    assert str(mode).lower() == "wal"
@@ -0,0 +1,205 @@
+from __future__ import annotations
+
+import json
+import math
+import os
+from collections.abc import Sequence
+from dataclasses import dataclass
+from typing import Any
+
+import httpx
+import pytest
+from pydantic_ai import Embedder
+from pydantic_ai.providers.openai import OpenAIProvider
+
+from stirling.documents.embedder import _build_embedder
+from stirling.documents.voyage import VOYAGE_BASE_URL, VoyageEmbeddingModel, build_voyage_model
+
+# Voyage's documented response body: OpenAI's shape, minus prompt_tokens.
+VOYAGE_RESPONSE = {
+    "object": "list",
+    "data": [
+        {"object": "embedding", "embedding": [0.1, 0.2, 0.3], "index": 0},
+        {"object": "embedding", "embedding": [0.4, 0.5, 0.6], "index": 1},
+    ],
+    "model": "voyage-4",
+    "usage": {"total_tokens": 7},
+}
+
+
+@dataclass
+class SentRequest:
+    url: str
+    auth: str | None
+    body: dict[str, Any]
+
+
+def _recording_model(sent: list[SentRequest]) -> VoyageEmbeddingModel:
+
+    def handler(request: httpx.Request) -> httpx.Response:
+        sent.append(
+            SentRequest(
+                url=str(request.url),
+                auth=request.headers.get("authorization"),
+                body=json.loads(request.content),
+            )
+        )
+        return httpx.Response(200, json=VOYAGE_RESPONSE)
+
+    client = httpx.AsyncClient(transport=httpx.MockTransport(handler))
+    provider = OpenAIProvider(base_url=VOYAGE_BASE_URL, api_key="pa-test-key", http_client=client)
+    return VoyageEmbeddingModel("voyage-4", provider=provider)
+
+
+@pytest.mark.anyio
+async def test_posts_to_voyage_embeddings_endpoint_with_bearer_auth() -> None:
+    sent: list[SentRequest] = []
+    await Embedder(_recording_model(sent)).embed_documents(["alpha", "beta"])
+
+    assert sent[0].url == f"{VOYAGE_BASE_URL}/embeddings"
+    assert sent[0].auth == "Bearer pa-test-key"
+    assert sent[0].body["model"] == "voyage-4"
+    assert sent[0].body["input"] == ["alpha", "beta"]
+
+
+@pytest.mark.anyio
+@pytest.mark.parametrize(
+    ("call", "expected"),
+    [("embed_query", "query"), ("embed_documents", "document")],
+)
+async def test_forwards_voyage_input_type(call: str, expected: str) -> None:
+    """The stock OpenAI model drops this field; Voyage needs it."""
+    sent: list[SentRequest] = []
+    embedder = Embedder(_recording_model(sent))
+    await getattr(embedder, call)(["text"])
+
+    assert sent[0].body["input_type"] == expected
+
+
+@pytest.mark.anyio
+async def test_caller_settings_win_over_the_default_input_type() -> None:
+    sent: list[SentRequest] = []
+    await Embedder(_recording_model(sent)).embed_documents(
+        ["text"], settings={"extra_body": {"input_type": "query", "output_dimension": 512}}
+    )
+
+    assert sent[0].body["input_type"] == "query"
+    assert sent[0].body["output_dimension"] == 512
+
+
+@pytest.mark.anyio
+async def test_parses_voyage_response_into_embeddings() -> None:
+    result = await Embedder(_recording_model([])).embed_documents(["alpha", "beta"])
+
+    assert result.embeddings == [[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]
+
+
+def test_build_voyage_model_reads_the_api_key_from_the_environment(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setenv("VOYAGE_API_KEY", "pa-env-key")
+
+    assert build_voyage_model("voyage-4").model_name == "voyage-4"
+
+
+def test_build_voyage_model_without_a_key_still_constructs(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.delenv("VOYAGE_API_KEY", raising=False)
+
+    assert build_voyage_model("voyage-4").model_name == "voyage-4"
+
+
+@pytest.mark.anyio
+async def test_embedding_without_a_key_fails_with_a_clear_error(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.delenv("VOYAGE_API_KEY", raising=False)
+    model = build_voyage_model("voyage-4")
+
+    with pytest.raises(ValueError, match="VoyageAI embeddings need an API key"):
+        await Embedder(model).embed_documents(["text"])
+
+
+@pytest.mark.anyio
+async def test_an_openai_key_is_never_sent_to_voyage(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.delenv("VOYAGE_API_KEY", raising=False)
+    monkeypatch.setenv("OPENAI_API_KEY", "sk-openai-secret")
+    model = build_voyage_model("voyage-4")
+
+    with pytest.raises(ValueError, match="VoyageAI embeddings need an API key"):
+        await Embedder(model).embed_documents(["text"])
+
+
+def test_env_form_routes_voyageai_through_the_adapter(monkeypatch: pytest.MonkeyPatch) -> None:
+    monkeypatch.setenv("VOYAGE_API_KEY", "pa-env-key")
+
+    embedder = _build_embedder("voyageai:voyage-4")
+
+    assert isinstance(embedder.model, VoyageEmbeddingModel)
+    assert embedder.model.model_name == "voyage-4"
+
+
+def test_config_push_form_routes_voyageai_through_the_adapter() -> None:
+    embedder = _build_embedder("voyage-4", provider="voyageai", api_key="pa-pushed-key")
+
+    assert isinstance(embedder.model, VoyageEmbeddingModel)
+
+
+def test_the_voyageai_sdk_is_not_installed() -> None:
+    """Guards the ~207MB the SDK would add back."""
+    with pytest.raises(ImportError):
+        __import__("voyageai")
+
+
+# Live checks, skipped unless VOYAGE_API_KEY is set so CI stays offline.
+live_only = pytest.mark.skipif(
+    not os.environ.get("VOYAGE_API_KEY"),
+    reason="set VOYAGE_API_KEY to run the live VoyageAI checks",
+)
+
+
+def _cosine(a: Sequence[float], b: Sequence[float]) -> float:
+    dot = sum(x * y for x, y in zip(a, b, strict=True))
+    return dot / (math.sqrt(sum(x * x for x in a)) * math.sqrt(sum(y * y for y in b)))
+
+
+@live_only
+@pytest.mark.anyio
+async def test_live_voyage_returns_usable_embeddings() -> None:
+    result = await Embedder(build_voyage_model("voyage-4")).embed_documents(["alpha", "beta"])
+
+    assert len(result.embeddings) == 2
+    assert len(result.embeddings[0]) == 1024
+
+
+@live_only
+@pytest.mark.anyio
+async def test_live_voyage_honours_input_type_server_side() -> None:
+    """Voyage embeds the same text differently per input_type."""
+    embedder = Embedder(build_voyage_model("voyage-4"))
+    text = "How do I combine two PDFs?"
+
+    as_query = await embedder.embed_query(text)
+    as_document = await embedder.embed_documents([text])
+
+    assert _cosine(as_query.embeddings[0], as_document.embeddings[0]) < 0.999
+
+
+@live_only
+@pytest.mark.anyio
+async def test_live_voyage_ranks_the_relevant_document_first() -> None:
+    embedder = Embedder(build_voyage_model("voyage-4"))
+    docs = await embedder.embed_documents(
+        ["Stirling PDF merges and splits PDF files.", "The capital of France is Paris."]
+    )
+    query = await embedder.embed_query("How do I combine two PDFs?")
+
+    relevant = _cosine(query.embeddings[0], docs.embeddings[0])
+    irrelevant = _cosine(query.embeddings[0], docs.embeddings[1])
+    assert relevant > irrelevant
+
+
+@live_only
+@pytest.mark.anyio
+async def test_live_voyage_accepts_voyage_only_parameters() -> None:
+    """output_dimension has no OpenAI equivalent, so this proves extra_body lands."""
+    result = await Embedder(build_voyage_model("voyage-4")).embed_documents(
+        ["dimension test"], settings={"extra_body": {"output_dimension": 256}}
+    )
+
+    assert len(result.embeddings[0]) == 256
@@ -0,0 +1,33 @@
+{
+  "printWidth": 80,
+  "tabWidth": 2,
+  "useTabs": false,
+  "endOfLine": "lf",
+  "sortPackageJson": false,
+  "ignorePatterns": [
+    "dist/",
+    "editor/dist/",
+    "editor/src-tauri/**/target/",
+    "editor/src-tauri/gen/",
+    "node_modules/",
+    "editor/public/vendor/",
+    "editor/public/mockServiceWorker.js",
+    "editor/public/og-metadata.json",
+    "editor/public/og-metadata.saas.json",
+    "editor/src/core/data/ogImageMap.json",
+    "editor/src/portal/generated/docsManifest.json",
+    "editor/public/pdfjs*/",
+    "editor/public/js/thirdParty/",
+    "editor/public/css/cookieconsent.css",
+    "storybook-static/",
+    "playwright-report/",
+    "editor/playwright-report/",
+    "test-results/",
+    "editor/test-results/",
+    "*.min.*",
+    "*.md",
+    "*.wxs",
+    "*.toml",
+    "editor/src/output.css"
+  ]
+}
@@ -1,30 +0,0 @@
-dist/
-editor/dist/
-# Tauri/Cargo build output (binary assets named *.js etc. confuse Prettier).
-# Match nested target/ dirs too - provisioner/ and thumbnail-handler/ each
-# have their own Cargo workspace under src-tauri/.
-editor/src-tauri/**/target/
-editor/src-tauri/gen/
-node_modules/
-editor/public/vendor/
-# Auto-generated by MSW (`msw init`); regenerated verbatim, not hand-formatted.
-editor/public/mockServiceWorker.js
-# Auto-generated OG/social-preview metadata (scripts/generate-og-metadata.mjs); regenerated verbatim.
-editor/public/og-metadata.json
-editor/public/og-metadata.saas.json
-editor/src/core/data/ogImageMap.json
-# Auto-generated portal docs manifest (scripts/sync-portal-docs.mts); regenerated verbatim.
-editor/src/portal/generated/docsManifest.json
-editor/public/pdfjs*/
-editor/public/js/thirdParty/
-editor/public/css/cookieconsent.css
-# Build / test artifacts that may exist locally even though they're gitignored
-storybook-static/
-playwright-report/
-editor/playwright-report/
-test-results/
-editor/test-results/
-*.min.*
-*.md
-*.wxs
-editor/src/output.css
@@ -1,6 +0,0 @@
-{
-  "printWidth": 80,
-  "tabWidth": 2,
-  "useTabs": false,
-  "endOfLine": "lf"
-}
@@ -18,7 +18,7 @@ For desktop app development, see the [Tauri](#tauri) section below.
 `frontend/` is a workspace containing one or more apps. Today it holds the
 PDF editor under `frontend/editor/`; new apps (the developer portal, etc.)
 will sit alongside it as siblings. Shared tooling — `package.json`, `node_modules`,
-`.storybook/`, oxlint, Prettier — lives at `frontend/` so every app installs
+`.storybook/`, oxlint, oxfmt — lives at `frontend/` so every app installs
 once and lints with the same config.

 ## Environment Variables
@@ -4592,6 +4592,10 @@ desc = "Change document restrictions and permissions"
 tags = "permissions,restrictions,rights,access control,allow,deny,printing,copying,editing,modify permissions,security settings,user rights"
 title = "Change Permissions"

+[home.classify]
+desc = "Identify what kind of document this is and tag it."
+title = "Classify"
+
 [home.compare]
 desc = "Compares and shows the differences between 2 PDF Documents"
 tags = "difference,compare,diff,compare PDFs,compare documents,find differences,show differences,changes,what changed,track changes,revisions,version compare,side by side,contrast,delta"
@@ -11044,6 +11048,7 @@ searchPlaceholder = "Search tools..."

 [toolPicker.subcategories]
 advancedFormatting = "Advanced Formatting"
+ai = "AI"
 automation = "Automation"
 developerTools = "Developer Tools"
 documentReview = "Document Review"
@@ -195,6 +195,11 @@
      "title": "Compress - Stirling PDF",
      "description": "Compress PDFs to reduce their file size."
    },
+    "classify": {
+      "image": "/og_images/home.png",
+      "title": "Classify - Stirling PDF",
+      "description": "Identify what kind of document this is and tag it."
+    },
    "extractPages": {
      "image": "/og_images/extract-pages.png",
      "title": "Extract Pages - Stirling PDF",
@@ -575,6 +580,7 @@
    "/remove-cert-sign": "removeCertSign",
    "/unlock-p-d-f-forms": "unlockPDFForms",
    "/compress": "compress",
+    "/classify": "classify",
    "/extract-pages": "extractPages",
    "/reorganize-pages": "reorganizePages",
    "/extract-images": "extractImages",
@@ -196,6 +196,11 @@
      "title": "Compress - Stirling PDF",
      "description": "Compress PDFs to reduce their file size."
    },
+    "classify": {
+      "image": "/og_images/home.png",
+      "title": "Classify - Stirling PDF",
+      "description": "Identify what kind of document this is and tag it."
+    },
    "extractPages": {
      "image": "/og_images/extract-pages.png",
      "title": "Extract Pages - Stirling PDF",
@@ -588,6 +593,7 @@
    "/remove-cert-sign": "removeCertSign",
    "/unlock-p-d-f-forms": "unlockPDFForms",
    "/compress": "compress",
+    "/classify": "classify",
    "/extract-pages": "extractPages",
    "/reorganize-pages": "reorganizePages",
    "/extract-images": "extractImages",
@@ -10,15 +10,9 @@ import { readFileSync, writeFileSync, mkdirSync } from "node:fs";
 import { dirname, resolve } from "node:path";
 import { parseArgs } from "node:util";
 import { compile, type JSONSchema } from "json-schema-to-typescript";
-import * as prettier from "prettier";

-// The API namespaces whose endpoints a pipeline can reference. `/api/v1/ai/tools/`
-// is absent from the spec, so it cannot appear here. Extend this list when other
-// namespaces become tools.
-//
-// `/api/v1/filter/` and `/api/v1/integration/` are included even though neither is a
-// user-facing tool: a stored pipeline can contain one, and ToolEndpoint keys the I/O
-// table, so leaving them out would stop a chain being checked past such a step.
+// Endpoints a pipeline can reference. filter/integration are not user-facing tools but a stored
+// pipeline can contain one; the AI namespace is admitted one endpoint at a time, not wholesale.
 const ALLOWED_PATH_PREFIXES = [
  "/api/v1/general/",
  "/api/v1/misc/",
@@ -26,6 +20,7 @@ const ALLOWED_PATH_PREFIXES = [
  "/api/v1/convert/",
  "/api/v1/filter/",
  "/api/v1/integration/",
+  "/api/v1/ai/tools/classify-and-label",
 ];

 // File plumbing, not user parameters: `fileInput` and `file` are the uploaded primary document
@@ -227,11 +222,7 @@ function collectToolIO(
  return { table, dropped };
 }

-async function renderToolIO(
-  spec: Json,
-  table: Record<string, unknown>,
-  outputPath: string,
-): Promise<string> {
+function renderToolIO(spec: Json, table: Record<string, unknown>): string {
  if (Object.keys(table).length === 0) {
    throw new Error(
      `No ${IO_EXTENSION} declarations in the spec. The backend publishes these from @ToolIO; regenerate with 'task backend:swagger'.`,
@@ -312,33 +303,12 @@ export function toolIOFor(
 }
 `;

-  const prettierConfig = await prettier.resolveConfig(outputPath);
-  return prettier.format(body, { ...prettierConfig, parser: "typescript" });
+  return body;
 }

-/** In check mode, fail when the committed file is out of date. */
-function writeOrCheck(
-  outputPath: string,
-  formatted: string,
-  check: boolean,
-  task: string,
-): void {
-  if (check) {
-    let current = "";
-    try {
-      current = readFileSync(outputPath, "utf-8");
-    } catch {
-      // Missing file counts as out of date.
-    }
-    if (current !== formatted) {
-      throw new Error(
-        `${outputPath} is out of date. Run '${task}' and commit the result.`,
-      );
-    }
-    return;
-  }
+function writeOutput(outputPath: string, contents: string): void {
  mkdirSync(dirname(outputPath), { recursive: true });
-  writeFileSync(outputPath, formatted, "utf-8");
+  writeFileSync(outputPath, contents, "utf-8");
 }

 async function main(): Promise<void> {
@@ -347,12 +317,11 @@ async function main(): Promise<void> {
      spec: { type: "string" },
      output: { type: "string" },
      "io-output": { type: "string" },
-      check: { type: "boolean", default: false },
    },
  });
  if (!values.spec || !values.output || !values["io-output"]) {
    throw new Error(
-      "Usage: generate-tool-api-types.mts --spec <SwaggerDoc.json> --output <file.ts> --io-output <file.ts> [--check]",
+      "Usage: generate-tool-api-types.mts --spec <SwaggerDoc.json> --output <file.ts> --io-output <file.ts>",
    );
  }
  const specPath = resolve(values.spec);
@@ -478,14 +447,9 @@ async function main(): Promise<void> {
      `Dropped ${dropped.length} @ToolIO declaration(s) on paths that are not tool endpoints. Add the namespace to ALLOWED_PATH_PREFIXES if a pipeline can contain these steps:\n  ${dropped.join("\n  ")}`,
    );
  }
-  writeOrCheck(
-    ioOutputPath,
-    await renderToolIO(spec, ioDeclarations, ioOutputPath),
-    values.check ?? false,
-    "task frontend:tool-models",
-  );
+  writeOutput(ioOutputPath, renderToolIO(spec, ioDeclarations));
  console.log(
-    `${values.check ? "Up to date" : "Generated"}: ${Object.keys(ioDeclarations).length} tool I/O declarations.`,
+    `Generated ${Object.keys(ioDeclarations).length} tool I/O declarations.`,
  );

  // Transitively inline every referenced component into `definitions`, rewriting its refs too.
@@ -508,7 +472,6 @@ async function main(): Promise<void> {
    definitions,
    fileFieldsByClass,
    outputPath,
-    values.check ?? false,
    skipped,
  );
 }
@@ -518,7 +481,6 @@ async function compileAndWrite(
  definitions: Record<string, Json>,
  fileFieldsByClass: Record<string, string[]>,
  outputPath: string,
-  check: boolean,
  skipped: string[],
 ): Promise<void> {
  // json-schema-to-typescript only emits a named, exported interface per schema
@@ -597,17 +559,9 @@ async function compileAndWrite(
  ].join("\n");

  const body = `${FILE_HEADER}\n\n${models}\n\n${footer}\n`;
-  const prettierConfig = await prettier.resolveConfig(outputPath);
-  const formatted = await prettier.format(body, {
-    ...prettierConfig,
-    parser: "typescript",
-  });
-
-  writeOrCheck(outputPath, formatted, check, "task frontend:tool-models");
-  console.log(
-    `${check ? "Up to date" : "Generated"}: ${tools.length} tool endpoints.`,
-  );
-  if (!check && skipped.length > 0) {
+  writeOutput(outputPath, body);
+  console.log(`Generated ${tools.length} tool endpoints.`);
+  if (skipped.length > 0) {
    console.log(
      `Skipped ${skipped.length} POST endpoint(s) with no request body: ${skipped.join(", ")}`,
    );
@@ -153,7 +153,7 @@ const BulkShareModal: React.FC<BulkShareModalProps> = ({
      if (onShared) {
        await onShared();
      }
-    } catch (error: any) {
+    } catch (error: unknown) {
      console.error("Failed to generate share link:", error);
      setErrorMessage(
        t(
@@ -1,5 +1,6 @@
 import { Stack, Card, Text, Flex } from "@mantine/core";
 import { Tooltip } from "@app/components/shared/Tooltip";
+import { TooltipTip } from "@app/types/tips";
 import { useTranslation } from "react-i18next";

 export interface CardOption<T = string> {
@@ -7,14 +8,14 @@ export interface CardOption<T = string> {
  prefixKey: string;
  nameKey: string;
  tooltipKey?: string;
-  tooltipContent?: any[];
+  tooltipContent?: TooltipTip[];
 }

 export interface CardSelectorProps<T, K extends CardOption<T>> {
  options: K[];
  onSelect: (value: T) => void;
  disabled?: boolean;
-  getTooltipContent?: (option: K) => any[];
+  getTooltipContent?: (option: K) => TooltipTip[];
 }

 const CardSelector = <T, K extends CardOption<T>>({
@@ -269,7 +269,7 @@ export function FileSelectorPicker({
              responseType: "blob",
              suppressErrorToast: true,
              skipAuthRedirect: true,
-            } as any,
+            },
          );
          const ct = readResponseHeader(res.headers, "content-type");
          const disp = readResponseHeader(res.headers, "content-disposition");
@@ -287,7 +287,7 @@ export function FileSelectorPicker({
              responseType: "blob",
              suppressErrorToast: true,
              skipAuthRedirect: true,
-            } as any,
+            },
          );
          const ct = readResponseHeader(res.headers, "content-type");
          const disp = readResponseHeader(res.headers, "content-disposition");
@@ -208,7 +208,7 @@ const FileSidebar = forwardRef<HTMLDivElement, FileSidebarProps>(
    const openWatchedFolders = useCallback(() => {
      if (collapsed && onToggleCollapse) onToggleCollapse();
      setCustomWorkbenchViewData(WATCHED_FOLDER_VIEW_ID, { folderId: null });
-      navActions.setWorkbench(WATCHED_FOLDER_WORKBENCH_ID as any);
+      navActions.setWorkbench(WATCHED_FOLDER_WORKBENCH_ID);
    }, [collapsed, onToggleCollapse, setCustomWorkbenchViewData, navActions]);

    // Clicking a file's membership dot jumps straight into that folder.
@@ -216,7 +216,7 @@ const FileSidebar = forwardRef<HTMLDivElement, FileSidebarProps>(
      (folderId: string) => {
        if (collapsed && onToggleCollapse) onToggleCollapse();
        setCustomWorkbenchViewData(WATCHED_FOLDER_VIEW_ID, { folderId });
-        navActions.setWorkbench(WATCHED_FOLDER_WORKBENCH_ID as any);
+        navActions.setWorkbench(WATCHED_FOLDER_WORKBENCH_ID);
      },
      [collapsed, onToggleCollapse, setCustomWorkbenchViewData, navActions],
    );
@@ -1,4 +1,5 @@
 import { useState } from "react";
+import axios from "axios";
 import { Modal, Stack, Text, PasswordInput, Alert } from "@mantine/core";
 import { Button } from "@app/ui/Button";
 import { useTranslation } from "react-i18next";
@@ -95,10 +96,13 @@ export default function FirstLoginModal({
      setTimeout(() => {
        onPasswordChanged();
      }, 1500);
-    } catch (err: any) {
+    } catch (err: unknown) {
      console.error("Failed to change password:", err);
+      const message = axios.isAxiosError<{ message?: string }>(err)
+        ? err.response?.data?.message
+        : undefined;
      setError(
-        err.response?.data?.message ||
+        message ||
          t(
            "firstLogin.passwordChangeFailed",
            "Failed to change password. Please check your current password.",
@@ -1,4 +1,4 @@
-import React, { CSSProperties, useMemo, useRef } from "react";
+import React, { CSSProperties, useCallback, useMemo, useRef } from "react";
 import { useAdjustFontSizeToFit } from "@app/components/shared/fitText/textFit";

 type FitTextProps = {
@@ -28,8 +28,14 @@ const FitText: React.FC<FitTextProps> = ({
 }) => {
  const ref = useRef<HTMLElement | null>(null);

+  // Callback ref: an HTMLElement handler satisfies span's/div's differing ref
+  // types (callback refs are contravariant), so the tag can stay polymorphic.
+  const setRef = useCallback((node: HTMLElement | null) => {
+    ref.current = node;
+  }, []);
+
  // Hook runs after mount and on size/text changes; uses observers internally
-  useAdjustFontSizeToFit(ref as any, {
+  useAdjustFontSizeToFit(ref, {
    maxFontSizePx: fontSize,
    minFontScale: minimumFontScale,
    maxLines: lines,
@@ -38,7 +44,7 @@ const FitText: React.FC<FitTextProps> = ({

  // Memoize the HTML tag to render (span/div) from the `as` prop so
  // React doesn't create a new component function on each render.
-  const ElementTag: any = useMemo(() => as, [as]);
+  const ElementTag: React.ElementType = useMemo(() => as, [as]);

  // For the / character, insert zero-width soft breaks to prefer wrapping at them
  const displayText = useMemo(() => {
@@ -70,7 +76,7 @@ const FitText: React.FC<FitTextProps> = ({

  return (
    <ElementTag
-      ref={ref}
+      ref={setRef}
      className={className}
      style={{ ...clampStyles, ...style }}
    >
@@ -165,7 +165,7 @@ const ShareFileModal: React.FC<ShareFileModalProps> = ({
      if (onUploaded) {
        await onUploaded();
      }
-    } catch (error: any) {
+    } catch (error: unknown) {
      console.error("Failed to generate share link:", error);
      setErrorMessage(
        t(
@@ -226,7 +226,7 @@ const ShareManagementModal: React.FC<ShareManagementModalProps> = ({
          durationMs: 2500,
        });
      }
-    } catch (error: any) {
+    } catch (error: unknown) {
      console.error("Failed to create share link:", error);
      setErrorMessage(
        t(
@@ -18,6 +18,18 @@ import { useLogoAssets } from "@app/hooks/useLogoAssets";
 import styles from "@app/components/shared/tooltip/Tooltip.module.css";
 import { Z_INDEX_OVER_FULLSCREEN_SURFACE } from "@app/styles/zIndex";

+// The wrapped child's own event handlers, which Tooltip forwards to after
+// running its own trigger logic. Kept partial since any given child may set none.
+interface ForwardedHandlers {
+  onPointerEnter?: React.PointerEventHandler;
+  onPointerLeave?: React.PointerEventHandler;
+  onMouseDown?: React.MouseEventHandler;
+  onMouseUp?: React.MouseEventHandler;
+  onClick?: React.MouseEventHandler;
+  onFocus?: React.FocusEventHandler;
+  onBlur?: React.FocusEventHandler;
+}
+
 export interface TooltipProps {
  sidebarTooltip?: boolean;
  position?: "right" | "left" | "top" | "bottom";
@@ -218,7 +230,7 @@ export const Tooltip: React.FC<TooltipProps> = ({
  const handlePointerEnter = useCallback(
    (e: React.PointerEvent) => {
      if (!isPinned && !disabled) openWithDelay();
-      (children.props as any)?.onPointerEnter?.(e);
+      (children.props as ForwardedHandlers).onPointerEnter?.(e);
    },
    [isPinned, openWithDelay, children.props, disabled],
  );
@@ -233,19 +245,19 @@ export const Tooltip: React.FC<TooltipProps> = ({
        tooltipRef.current &&
        tooltipRef.current.contains(related)
      ) {
-        (children.props as any)?.onPointerLeave?.(e);
+        (children.props as ForwardedHandlers).onPointerLeave?.(e);
        return;
      }

      // Ignore transient leave between mousedown and click
      if (clickPendingRef.current) {
-        (children.props as any)?.onPointerLeave?.(e);
+        (children.props as ForwardedHandlers).onPointerLeave?.(e);
        return;
      }

      clearTimers();
      if (allowAutoClose && !isPinned) setOpen(false);
-      (children.props as any)?.onPointerLeave?.(e);
+      (children.props as ForwardedHandlers).onPointerLeave?.(e);
    },
    [clearTimers, isPinned, setOpen, children.props, allowAutoClose],
  );
@@ -253,7 +265,7 @@ export const Tooltip: React.FC<TooltipProps> = ({
  const handleMouseDown = useCallback(
    (e: React.MouseEvent) => {
      clickPendingRef.current = true;
-      (children.props as any)?.onMouseDown?.(e);
+      (children.props as ForwardedHandlers).onMouseDown?.(e);
    },
    [children.props],
  );
@@ -262,7 +274,7 @@ export const Tooltip: React.FC<TooltipProps> = ({
    (e: React.MouseEvent) => {
      // allow microtask turn so click can see this false
      queueMicrotask(() => (clickPendingRef.current = false));
-      (children.props as any)?.onMouseUp?.(e);
+      (children.props as ForwardedHandlers).onMouseUp?.(e);
    },
    [children.props],
  );
@@ -279,7 +291,7 @@ export const Tooltip: React.FC<TooltipProps> = ({
        return;
      }
      clickPendingRef.current = false;
-      (children.props as any)?.onClick?.(e);
+      (children.props as ForwardedHandlers).onClick?.(e);
    },
    [clearTimers, pinOnClick, open, setOpen, children.props],
  );
@@ -288,7 +300,7 @@ export const Tooltip: React.FC<TooltipProps> = ({
  const handleFocus = useCallback(
    (e: React.FocusEvent) => {
      if (!isPinned && !disabled && openOnFocus) openWithDelay();
-      (children.props as any)?.onFocus?.(e);
+      (children.props as ForwardedHandlers).onFocus?.(e);
    },
    [isPinned, openWithDelay, children.props, disabled, openOnFocus],
  );
@@ -301,12 +313,12 @@ export const Tooltip: React.FC<TooltipProps> = ({
        tooltipRef.current &&
        tooltipRef.current.contains(related)
      ) {
-        (children.props as any)?.onBlur?.(e);
+        (children.props as ForwardedHandlers).onBlur?.(e);
        return;
      }
      clearTimers();
      if (allowAutoClose && !isPinned) setOpen(false);
-      (children.props as any)?.onBlur?.(e);
+      (children.props as ForwardedHandlers).onBlur?.(e);
    },
    [isPinned, setOpen, children.props, allowAutoClose, clearTimers],
  );
@@ -339,24 +351,30 @@ export const Tooltip: React.FC<TooltipProps> = ({
  );

  // Enhance child with handlers and ref
-  const childWithHandlers = React.cloneElement(children as any, {
-    ref: (node: HTMLElement | null) => {
-      triggerRef.current = node || null;
-      const originalRef = (children as any).ref;
-      if (typeof originalRef === "function") originalRef(node);
-      else if (originalRef && typeof originalRef === "object")
-        (originalRef as any).current = node;
+  const childWithHandlers = React.cloneElement(
+    children as React.ReactElement<Record<string, unknown>>,
+    {
+      ref: (node: HTMLElement | null) => {
+        triggerRef.current = node || null;
+        const originalRef = (
+          children as React.ReactElement & { ref?: React.Ref<HTMLElement> }
+        ).ref;
+        if (typeof originalRef === "function") originalRef(node);
+        else if (originalRef && typeof originalRef === "object")
+          (originalRef as React.MutableRefObject<HTMLElement | null>).current =
+            node;
+      },
+      "aria-describedby": open ? tooltipIdRef.current : undefined,
+      onPointerEnter: handlePointerEnter,
+      onPointerLeave: handlePointerLeave,
+      onMouseDown: handleMouseDown,
+      onMouseUp: handleMouseUp,
+      onClick: handleClick,
+      onFocus: handleFocus,
+      onBlur: handleBlur,
+      onKeyDown: handleKeyDown,
    },
-    "aria-describedby": open ? tooltipIdRef.current : undefined,
-    onPointerEnter: handlePointerEnter,
-    onPointerLeave: handlePointerLeave,
-    onMouseDown: handleMouseDown,
-    onMouseUp: handleMouseUp,
-    onClick: handleClick,
-    onFocus: handleFocus,
-    onBlur: handleBlur,
-    onKeyDown: handleKeyDown,
-  });
+  );

  const shouldShowTooltip = open;
  const shouldShowCloseButton = showCloseButton || isPinned;
@@ -25,6 +25,13 @@ export default function WorkbenchBarMobileActions({
 }: WorkbenchBarActionsProps) {
  const { t } = useTranslation();
  const exportDisabled = actionsDisabled || policyEnforcing;
+  const showPrint = currentView === "viewer";
+  const showFileActions = !isCustomView;
+
+  // Custom workbench views own their content, so none of these apply. The
+  // desktop cluster renders nothing at all in that case; without this the
+  // trigger would still be there, opening an empty dropdown.
+  if (!showPrint && !showFileActions) return null;

  return (
    <Menu shadow="md" width={230} position="bottom-end">
@@ -39,7 +46,7 @@ export default function WorkbenchBarMobileActions({
        </ActionIcon>
      </Menu.Target>
      <Menu.Dropdown>
-        {currentView === "viewer" && (
+        {showPrint && (
          <Menu.Item
            leftSection={<PrintIcon sx={{ fontSize: "1.1rem" }} />}
            disabled={exportDisabled}
@@ -48,7 +55,7 @@ export default function WorkbenchBarMobileActions({
            {t("workbenchBar.print", "Print PDF")}
          </Menu.Item>
        )}
-        {!isCustomView && (
+        {showFileActions && (
          <Menu.Item
            leftSection={
              <LocalIcon
@@ -63,7 +70,7 @@ export default function WorkbenchBarMobileActions({
            {downloadLabel}
          </Menu.Item>
        )}
-        {!isCustomView && saveAsIconName && (
+        {showFileActions && saveAsIconName && (
          <Menu.Item
            leftSection={
              <LocalIcon icon={saveAsIconName} width="1.1rem" height="1.1rem" />
@@ -74,7 +81,7 @@ export default function WorkbenchBarMobileActions({
            {t("workbenchBar.saveAs", "Save As")}
          </Menu.Item>
        )}
-        {!isCustomView && (
+        {showFileActions && (
          <>
            <Menu.Divider />
            <Menu.Item
@@ -282,6 +282,9 @@ export default function RightSidebar() {
              onShowAllTools={handleShowAllTools}
              onToolSelect={handleToolSelectWithTransition}
              compact={false}
+              /* Mobile keeps the workbench bar - and with it the super search -
+                 on the other slide, so the list needs its own filter. */
+              showSearch={isMobile}
            />
          </>
        </div>
@@ -183,6 +183,17 @@
  }
 }

+/* In-panel tool filter. Aligned with .tool-picker__compact's inline padding so
+   the field lines up with the tool rows underneath it. */
+.tool-panel__search {
+  flex-shrink: 0;
+  padding: 0.5rem var(--mantine-spacing-sm) 0;
+}
+
+.tool-panel__search .search-input-container {
+  margin: 0;
+}
+
 .tool-panel__compact-header-actions {
  display: flex;
  align-items: center;
@@ -4,6 +4,7 @@ import { useToolWorkflow } from "@app/contexts/ToolWorkflowContext";
 import ToolPicker from "@app/components/tools/ToolPicker";
 import SearchResults from "@app/components/tools/SearchResults";
 import ToolRenderer from "@app/components/tools/ToolRenderer";
+import ToolSearch from "@app/components/tools/toolPicker/ToolSearch";
 import { ToolPanelViewerBar } from "@app/components/tools/ToolPanelViewerBar";
 import { ToolId } from "@app/types/toolId";

@@ -20,6 +21,11 @@ interface ToolPanelProps {
  onToolSelect?: (id: ToolId) => void;
  /** Whether to render the compact (favourites + recommended only) view. */
  compact?: boolean;
+  /**
+   * Render a tool filter at the head of the panel. Set where the workbench
+   * bar's super search is out of reach, so the list stays searchable in place.
+   */
+  showSearch?: boolean;
 }

 /** Tool list and renderer for the right rail; rail chrome lives in RightSidebar. */
@@ -28,24 +34,44 @@ export default function ToolPanel({
  onShowAllTools,
  onToolSelect,
  compact: compactProp,
+  showSearch = false,
 }: ToolPanelProps) {
  const { t } = useTranslation();
  const {
    leftPanelView,
    searchQuery,
+    setSearchQuery,
    filteredTools,
+    toolRegistry,
    selectedToolKey,
    handleToolSelect,
    setPreviewFile,
  } = useToolWorkflow();
  const selectTool = onToolSelect ?? handleToolSelect;

+  // Only offer the filter over the list itself; once a tool is open the panel
+  // belongs to that tool. Deriving the results branch from the same flag keeps
+  // the input and what it filters from drifting apart.
+  const panelSearch = showSearch && leftPanelView === "toolPicker";
+  const searching = searchQuery.trim().length > 0;
+
  return (
    <>
      {/* Viewer mode tools — annotate, redact, form fill */}
      <ToolPanelViewerBar />

-      {allToolsView && searchQuery.trim().length > 0 ? (
+      {panelSearch && (
+        <div className="tool-panel__search">
+          <ToolSearch
+            value={searchQuery}
+            onChange={setSearchQuery}
+            toolRegistry={toolRegistry}
+            mode="filter"
+          />
+        </div>
+      )}
+
+      {searching && (allToolsView || panelSearch) ? (
        <div className="flex-1 flex flex-col overflow-y-auto">
          <SearchResults
            filteredTools={filteredTools}
@@ -59,7 +85,7 @@ export default function ToolPanel({
            selectedToolKey={selectedToolKey}
            onSelect={(id) => selectTool(id as ToolId)}
            filteredTools={filteredTools}
-            isSearching={Boolean(searchQuery && searchQuery.trim().length > 0)}
+            isSearching={searching}
            compact={compactProp ?? !allToolsView}
            onShowAllTools={onShowAllTools}
          />
@@ -82,7 +82,7 @@ export default function AutomationCreation({
    setConfigModalOpen(true);
  };

-  const handleToolConfigSave = (parameters: Record<string, any>) => {
+  const handleToolConfigSave = (parameters: Record<string, unknown>) => {
    if (configuraingToolIndex >= 0) {
      updateTool(configuraingToolIndex, {
        configured: true,
@@ -19,7 +19,7 @@ interface AutomationEntryProps {
  /** Optional description for tooltip */
  description?: string;
  /** MUI Icon component for the badge */
-  badgeIcon?: React.ComponentType<any>;
+  badgeIcon?: React.ComponentType;
  /** Array of tool operation names in the workflow */
  operations: string[];
  /** Click handler */
@@ -9,11 +9,12 @@ import { useToolRegistry } from "@app/contexts/ToolRegistryContext";
 import { AutomationConfig, ExecutionStep } from "@app/types/automation";
 import { EXECUTION_STATUS } from "@app/constants/automation";
 import { useResourceCleanup } from "@app/utils/resourceManager";
+import type { useAutomateOperation } from "@app/hooks/tools/automate/useAutomateOperation";

 interface AutomationRunProps {
  automation: AutomationConfig;
  onComplete: () => void;
-  automateOperation?: any; // TODO: Type this properly when available
+  automateOperation?: ReturnType<typeof useAutomateOperation>;
 }

 export default function AutomationRun({
@@ -34,13 +35,13 @@ export default function AutomationRun({
  // Use the operation hook's loading state
  const isExecuting = automateOperation?.isLoading || false;
  const hasResults =
-    automateOperation?.files.length > 0 ||
+    (automateOperation?.files.length ?? 0) > 0 ||
    automateOperation?.downloadUrl !== null;

  // Initialize execution steps from automation
  useEffect(() => {
    if (automation?.operations) {
-      const steps = automation.operations.map((op: any, index: number) => {
+      const steps = automation.operations.map((op, index) => {
        const tool = toolRegistry[op.operation as keyof typeof toolRegistry];
        return {
          id: `${op.operation}-${index}`,
@@ -125,7 +126,7 @@ export default function AutomationRun({
      // Mark all as completed and reset current step
      setCurrentStepIndex(-1);
      console.log(`✅ Automation completed successfully`);
-    } catch (error: any) {
+    } catch (error: unknown) {
      console.error("Automation execution failed:", error);
      setCurrentStepIndex(-1);
    }
@@ -4,6 +4,7 @@ import { Stack, Text, ScrollArea } from "@mantine/core";
 import {
  ToolRegistryEntry,
  ToolRegistry,
+  SubcategoryId,
  getToolSupportsAutomate,
 } from "@app/data/toolsTaxonomy";
 import { useToolSections } from "@app/hooks/useToolSections";
@@ -81,9 +82,7 @@ export default function ToolSelector({
  }, [filteredTools]);

  // Use the same tool sections logic as the main ToolPicker
-  const { sections, searchGroups } = useToolSections(
-    transformedFilteredTools as any /* FIX ME */,
-  );
+  const { sections, searchGroups } = useToolSections(transformedFilteredTools);

  // Determine what to display: search results or organized sections
  const isSearching = searchTerm.trim().length > 0;
@@ -98,7 +97,9 @@ export default function ToolSelector({
        return [
          {
            name: "Tools",
-            subcategoryId: "all" as any,
+            // Synthetic "all tools" group used only as a fallback when the
+            // taxonomy produces no sections; "all" is not a real SubcategoryId.
+            subcategoryId: "all" as unknown as SubcategoryId,
            tools: baseFilteredTools.map(([key, tool]) => ({ id: key, tool })),
          },
        ];
@@ -125,7 +126,7 @@ export default function ToolSelector({
      displayGroups.map((subcategory) =>
        renderToolButtons(
          t,
-          subcategory as any,
+          subcategory,
          null,
          handleToolSelect,
          !isSearching,
@@ -13,6 +13,7 @@ import {
  useFileActions,
 } from "@app/contexts/FileContext";
 import { useFileWithUrl } from "@app/hooks/useFileWithUrl";
+import { ZoomMode } from "@embedpdf/plugin-zoom/react";
 import { useViewer } from "@app/contexts/ViewerContext";
 import { LocalEmbedPDF } from "@app/components/viewer/LocalEmbedPDF";
 import { PdfViewerToolbar } from "@app/components/viewer/PdfViewerToolbar";
@@ -418,7 +419,7 @@ const EmbedPdfViewerContent = ({
                return;
              case "0":
                event.preventDefault();
-                zoomActions.requestZoom("fit-width");
+                zoomActions.requestZoom(ZoomMode.FitWidth);
                return;
            }
          }
@@ -118,7 +118,6 @@ export function SelectionAPIBridge() {

    const buildApi = () => ({
      copyToClipboard: () => selection.copyToClipboard(),
-      getSelectedText: () => selection.getSelectedText(),
      getFormattedSelection: () => selection.getFormattedSelection(),
      selectAll: async (totalPages: number) => {
        const docId = activeDocumentId;
@@ -6,6 +6,8 @@ import {
  ZoomState,
 } from "@app/contexts/viewer/viewerBridges";
 import { PdfBookmarkObject, PdfAttachmentObject } from "@embedpdf/models";
+import { ZoomLevel, Point } from "@embedpdf/plugin-zoom";
+import { FormattedSelection } from "@embedpdf/plugin-selection";

 export interface ScrollActions {
  scrollToPage: (page: number, behavior?: "smooth" | "instant") => void;
@@ -19,7 +21,7 @@ export interface ZoomActions {
  zoomIn: () => void;
  zoomOut: () => void;
  toggleMarqueeZoom: () => void;
-  requestZoom: (level: any, center?: any) => void;
+  requestZoom: (level: ZoomLevel, center?: Point) => void;
  setZoomLevel: (factor: number) => void;
 }

@@ -31,8 +33,7 @@ export interface PanActions {

 export interface SelectionActions {
  copyToClipboard: () => void;
-  getSelectedText: () => string;
-  getFormattedSelection: () => any;
+  getFormattedSelection: () => FormattedSelection[] | null;
  selectAll: (totalPages: number) => Promise<boolean>;
  selectWordAt: (pageIndex: number, x: number, y: number) => boolean;
 }
@@ -51,7 +52,7 @@ export interface RotationActions {
 }

 export interface SearchActions {
-  search: (query: string) => Promise<any> | undefined;
+  search: (query: string) => Promise<unknown> | undefined;
  next: () => void;
  previous: () => void;
  clear: () => void;
@@ -214,7 +215,7 @@ export function createViewerActions({
        api.toggleMarqueeZoom();
      }
    },
-    requestZoom: (level: any, center?: any) => {
+    requestZoom: (level: ZoomLevel, center?: Point) => {
      const api = registry.current.zoom?.api;
      if (api?.requestZoom) {
        api.requestZoom(level, center);
@@ -257,13 +258,6 @@ export function createViewerActions({
        api.copyToClipboard();
      }
    },
-    getSelectedText: () => {
-      const api = registry.current.selection?.api;
-      if (api?.getSelectedText) {
-        return api.getSelectedText() ?? "";
-      }
-      return "";
-    },
    getFormattedSelection: () => {
      const api = registry.current.selection?.api;
      if (api?.getFormattedSelection) {
@@ -1,5 +1,7 @@
 import { SpreadMode } from "@embedpdf/plugin-spread/react";
 import { PdfBookmarkObject, PdfAttachmentObject } from "@embedpdf/models";
+import { ZoomLevel, Point } from "@embedpdf/plugin-zoom";
+import { FormattedSelection } from "@embedpdf/plugin-selection";

 export enum PdfPermissionFlag {
  Print = 0x0004,
@@ -46,7 +48,7 @@ export interface ZoomAPIWrapper {
  zoomIn: () => void;
  zoomOut: () => void;
  toggleMarqueeZoom: () => void;
-  requestZoom: (level: any, center?: any) => void;
+  requestZoom: (level: ZoomLevel, center?: Point) => void;
 }

 export interface PanAPIWrapper {
@@ -58,8 +60,7 @@ export interface PanAPIWrapper {

 export interface SelectionAPIWrapper {
  copyToClipboard: () => void;
-  getSelectedText: () => string | any;
-  getFormattedSelection: () => any;
+  getFormattedSelection: () => FormattedSelection[];
  selectAll: (totalPages: number) => Promise<boolean>;
  selectWordAt: (pageIndex: number, x: number, y: number) => boolean;
 }
@@ -79,7 +80,7 @@ export interface RotationAPIWrapper {
 }

 export interface SearchAPIWrapper {
-  search: (query: string) => Promise<any>;
+  search: (query: string) => Promise<unknown>;
  clear: () => void;
  next: () => void;
  previous: () => void;
@@ -0,0 +1,76 @@
+/**
+ * Classification is a thing a pipeline can do, not a thing only the Classification policy may do.
+ *
+ * The chain that used to stop it: `getExecutableTools` drops any tool whose endpoint is not a
+ * member of the generated `ToolEndpoint` union; that union comes from the OpenAPI spec, gated by
+ * the generator's namespace allowlist; and the classify controller was `@Hidden`, so it never
+ * reached the spec at all. These tests pin each link, because any one of them silently removes the
+ * step from the builder's picker rather than failing loudly.
+ */
+import { describe, expect, test, vi } from "vitest";
+import { renderHook } from "@testing-library/react";
+import { useTranslatedToolCatalog } from "@app/data/useTranslatedToolRegistry";
+import { getExecutableTools } from "@app/hooks/tools/shared/toolAutomation";
+import { isToolEndpoint } from "@app/hooks/tools/shared/toolApiMapping";
+import { TOOL_IO } from "@app/types/toolIO";
+import { filterToolRegistryByQuery } from "@app/utils/toolSearch";
+
+vi.mock("react-i18next", () => ({
+  useTranslation: () => ({
+    t: (key: string, fallback?: string) => fallback ?? key,
+    i18n: { changeLanguage: vi.fn(), language: "en-US" },
+  }),
+  Trans: ({ children }: { children?: unknown }) => children,
+}));
+
+const CLASSIFY_ENDPOINT = "/api/v1/ai/tools/classify-and-label";
+
+describe("classify as a pipeline task", () => {
+  test("the classify endpoint is a generated ToolEndpoint", () => {
+    // Fails if the controller goes back to @Hidden, or the generator's allowlist drops the
+    // /api/v1/ai/tools/ namespace, or nobody regenerated after either.
+    expect(isToolEndpoint(CLASSIFY_ENDPOINT)).toBe(true);
+  });
+
+  test("the builder offers it as a step", () => {
+    const { result } = renderHook(() => useTranslatedToolCatalog());
+
+    const executable = getExecutableTools(result.current.regularTools);
+    const classify = executable.find((tool) => tool.toolId === "classify");
+
+    expect(classify).toBeDefined();
+    expect(classify?.endpoint).toBe(CLASSIFY_ENDPOINT);
+  });
+
+  test("it declares PDF in, PDF out, so a chain can be checked across it", () => {
+    // Without this the builder shows "Can't check what this step accepts" and validation stops
+    // dead at the step - the I/O table is keyed by endpoint and comes from @ToolIO in the spec.
+    expect(TOOL_IO[CLASSIFY_ENDPOINT]).toEqual({
+      accepts: ["PDF"],
+      produces: "PDF",
+      arity: "SISO",
+    });
+  });
+
+  test("it is offered to pipelines but kept out of the editor's tool list", () => {
+    const { result } = renderHook(() => useTranslatedToolCatalog());
+
+    // There is no interactive classify tool to open - it only means something inside a pipeline.
+    expect(result.current.regularTools.classify?.hiddenFromToolList).toBe(true);
+    expect(
+      filterToolRegistryByQuery(result.current.regularTools, "").some(
+        (ranked) => ranked.item[0] === "classify",
+      ),
+    ).toBe(false);
+  });
+
+  test("it does not re-classify by default", () => {
+    const { result } = renderHook(() => useTranslatedToolCatalog());
+
+    const config = result.current.regularTools.classify?.operationConfig;
+
+    // The step is idempotent unless asked otherwise: a second run on a classified document
+    // would be a second engine call, and a second charge, for the same answer.
+    expect(config?.defaultParameters).toEqual({ reclassify: false });
+  });
+});
--- a/Show More
+++ b/Show More